Section Also Discuss Two Exotic Biological Interest to the Project

VII. Future Directions

Grasshoppers that die from exposure to fungi in the Entomophaga genus characteristi- cally do so in a heads-up position. The GHIPM Project studied the use of Entomophaga grylli in the hope of lessening overall dependence on chemical control to manage rangeland grasshopper populations. (APHIS file photo.)

VII.1 Future Directions in Grasshopper Management—An Introduction

Jerome A. Onsager

The following chapters about future directions in grass- Such a strategy, called inoculative release, appeals to hopper management fall into three general categories. some pest managers because the organisms could become These categories can be described as potential new agents self-perpetuating and therefore permanent deterrents to for grasshopper suppression, emerging new ecological grasshopper populations. Conversely, inoculative release information that could be integrated into grasshopper is worrisome to others because it could produce undesir- management systems, and issues that could affect grass- able side effects that also could become permanent. At hopper management priorities, especially on public lands. this time, it appears unlikely that current regulatory guidelines will allow the release of the two exotic agents. Chapters VII.2, 3, 5, 6, 7, and 8 discuss a number of promising new candidate agents for conventional short- Chapters VII.11, 13, 14, and 15 discuss areas of unfin- term control of economic infestations of grasshoppers. ished long-term research on grasshopper ecology. Hope- The agents’ eventual viability will be dictated primarily fully, the finished products someday will be incorporated by the same practical elements that affect current control into improved land-management systems. An under- tactics. These elements include standing of how grasshoppers respond to controllable • approval by regulatory agencies, attributes of habitat can be exploited in management • reasonable production costs plus economical volume systems that reduce the frequency and intensity of grass- for the producer, hopper depredation. • reasonable shelf life and consistency of demand for the distributor, and Finally, this handbook would be incomplete without • satisfaction plus reasonable profit for the consumer. some direct input into the complex and competing social, A candidate agent that is deficient in any critical element political, and environmental issues that affect grasshop- will not compete strongly with current technology until per management on public lands. Chapters VII.10, 12, the deficiency is corrected. and 16 are contributions that obviously are within the competence and responsibility of GHIPM and are of Chapters in this section also discuss two exotic biological interest to the Project. The information is intended not to control agents that were considered by the Federal Gov- provide definitive solutions to problems but rather to be ernment for nonconventional long-term suppression of available when conflicts of interest must be resolved. grasshopper populations. Grasshopper Integrated Pest Management (GHIPM) Project scientists evaluated a fungal pathogen (chapter VII.4) and an egg parasite (chapter VII.9) from Australia as candidates for release in the United States to build a reservoir of biological control.

VII.1–1

VII.2 Dimilin® Spray for Reducing Rangeland Grasshopper Populations

R. N. Foster and K. Christian Reuter

Introduction armyworm, and boll weevil on cotton, several insects on soybean, several forest pest insects, and in California on The insecticides used to control outbreaks of grasshop- mosquito larvae. Because of its mode of action, pers on rangeland are active against a broad spectrum of nonchitin-forming animals and adult insects and spiders insects, in both adult and immature stages. For rangeland enjoy a reduced risk compared to that of conventional use in Animal and Plant Health Inspection Service insecticides. (APHIS) cooperative programs, pest managers apply insecticides at doses and in formulations that have a Several studies have been conducted with Dimilin formu- minimal but acceptable impact on nontarget insects while lated into a bran-based bait for grasshoppers. Wang and substantially reducing grasshoppers. Because their activ- Fuller (1991) demonstrated the effectiveness of 1 and 2 ity is broad, these insecticidal sprays sometimes reduce lb of 1 percent diflubenzuron bran bait per acre against some nonpest insect species in the target areas. However, rangeland grasshoppers on 12-acre plots in southwestern populations of nontargets have been seen to rebound rela- South Dakota. Bomar and Lockwood (1991) demon- tively quickly following treatments on rangeland, even strated the effectiveness of the same formula and rate over large areas (see chapter III.3, “Impact of Control against rangeland grasshoppers on 10-acre plots in east- Programs on Nontarget Arthropods”). While undesir- ern Wyoming. Both of these studies utilized ground able, the effects of these sprays on nontarget insects are equipment for application. In two 2-year studies where acceptable. Short-term reductions in nontargets are part bait was aerially applied to replicated 40-acre plots, Jech of the price pest managers currently pay for artificially et al. (1993) showed diflubenzuron and carbaryl bran bait bringing an outbreak of grasshoppers back to a normal treatments to be equally effective on mixed populations level. of grasshoppers. (Figures VII.2–1 and –2 illustrate technical challenges in using bran materials in aerial spray The goals of insect control today are rapidly expanding. programs.) However, the study indicated that the species It is environmentally advantageous to reduce the minimal Phlibostroma quadrimaculatum (Thomas) could be con- effects of sprays on nontargets even further. Increasing trolled with diflubenzuron when not controlled with car- protection to nontargets, particularly those that naturally baryl bait. work to keep grasshopper populations in balance, supports basic integrated pest management (IPM) objectives Results of these studies are very promising. However, that encourage and emphasize the use of naturally occur- some damaging species of grasshopper do not readily ring organisms. accept the bran baits and may remain at undesirable levels (Jech et al. 1989 unpubl., 1992 unpubl., and 1993; Some insecticides, called insect growth regulators, have a Onsager et al. 1990; Quinn et al. 1989). Additionally, narrower spectrum of activity and cause death in a man- levels of reduction with all bran-based baits on suscep- ner different from most broad-spectrum insecticides. The tible species tend to be lower when compared to spray Dimilin® brand of diflubenzuron, (1-(4-clorophenyl)-3- treatments that are deposited directly on both the pest and (2,6 diflourobenzoyl)-urea, is one of these growth regula- the preferred food of the pest. tors. It inhibits chitin biosyntheses and thereby interferes with the formation and deposition of the chitin in the In an effort to take advantage of the desirable qualities of cuticle in an insect exoskeleton. This disruption of Dimilin while avoiding the general limitations of bran normal development may result in death to the insect baits, APHIS scientists at the Phoenix Methods Develop- when molting is attempted. ment Center studied spray formulations. Compared to currently used broad-spectrum insecticides, Dimilin Diflubenzuron has been shown to be effective against should lessen the impact on those nontarget insects and immature stages of several insect pests and is registered arachnids that are in an adult stage at the time the grass- in the United States for control of beet armyworm, fall hoppers are treated.

VII.2–1 Evaluating Potential Treatment Rates—A Field Study

In 1991 we conducted a detailed study to (1) generally evaluate an aerially sprayed formulation for control of grasshoppers on rangeland, (2) determine the most effective dose of three candidate doses for achieving immediate and seasonlong effectiveness on both the total grasshopper population and the individual component species of the population, and (3) determine the usefulness of the treatments for suppression or controlling migration into the treated area during the season of treatment.

In this study, we applied three doses of Dimilin 25W spray in volumes of 32 oz/acre to 40-acre mixed-grass rangeland plots in western South Dakota. Three sets of plots were treated with Dimilin spray at 0.015, 0.030, and 0.045 lb active ingredient (AI) per acre. An additional set of plots was sprayed with the standard carbaryl rangeland treatment (Sevin®-4 Oil ULV at 0.5 lb AI/acre) for comparison. A fifth set of plots was left untreated. When applications were made, most grasshoppers were in the second or third instar.

We found that all three dosages of Dimilin caused reduction as great as the standard carbaryl treatment after 1 week. After 2 weeks, all treatments showed reduction in the range of 94 to 96 percent. Reductions continued to increase to the end of the study and 9 weeks after treat- Figure VII.2–1—A load of bran is delivered for onsite mixing with ment ranged from 96 to 98 percent. chemicals or insect growth regulators at an airstrip in the Dakotas. (Agricultural Research Service photo by John Kucharsky.) Overall, we found no differences in the effects of Dimilin and carbaryl. Dimilin showed almost immediate acceptable reduction of grasshoppers within 7 days and continued to be effective throughout the season of treatment. Measurable migration into the Dimilin-treated plots was undetectable. Surviving hatch that might have occurred was also undetectable. In this study, in terms of providing acceptable control, Dimilin proved to be an excellent alternative for consideration when treating grasshoppers on rangeland.

VII.2–2 Figure VII.2–2—The treated bran bait is sacked and then dumped into a chamber in the fuselage of the spray plane. Inside that chamber, APHIS- developed aerating equipment keeps the bran bait from clumping, which would cause uneven applications of product. (Agricultural Research Service photo by John Kucharsky.)

Use of Dimilin Spray Under Operational We found that the standard (Sevin-4 Oil ULV) treatment Conditions caused greater reductions in grasshoppers after 1 week than the Dimilin treatments. After 2 weeks, all three In 1993, we conducted a study to evaluate the usefulness treatments caused reductions in grasshoppers that would of two formulations of Dimilin for control of grasshop- be acceptable in large-scale program efforts. However, pers on rangeland under operational conditions that could the Dimilin 2F and carbaryl treatments were causing be experienced during a large-scale grasshopper control greater reductions than the Dimilin 25W. Mortality at program. In this study, we aerially applied Dimilin 25W, 3 weeks after application showed that all three treatments Dimilin 2F, and carbaryl (Sevin-4 Oil ULV) to mixed- were performing equally well. After 4 weeks, we found grass rangeland plots in western North Dakota. All three that the Dimilin 2F formulation caused greater reductions formulations were sprayed in a diesel carrier. We applied in grasshoppers compared to the other treatments. each treatment to a square 640-acre block. Both Dimilin Trends in our study showed that mortality increased over treatments were applied at the dose of 0.0156 lb AI/acre the 4 weeks after treatment with Dimilin 2F and started to in 32 fluid oz of mix. The carbaryl treatment was applied decline with Dimilin 25W and Sevin-4 Oil ULV between in 20 fluid oz of mix per acre at the dose of 0.5 lb AI and the second and third week after treatment. was used as a standard rangeland treatment for comparison. We compared reduction in grasshopper populations From a cursory examination of the study area 16 weeks within the operational plots to populations of untreated after treatment, we found that no obvious additional grasshoppers in adjacent areas surrounding the treated hatch had survived, nor had any migration into the treated plots. Most grasshoppers treated were in the second or area occurred. Densities of grasshoppers were no greater third instar. than at 4 weeks after treatment.

VII.2–3 In our operational study, the 2F formulation proved to be Onsager, J. A.; Henry, J. E.; Foster, R. N.; Staten, R. T. 1980. Accep- more compatible with the spraying system. The 25W tance of wheat bran bait by species of rangeland grasshoppers. Journal formulation mixed with diesel resulted in a precipitant of Economic Entomology 73: 548–551. that could potentially cause a clogging problem with the Quinn, M. A.; Kepner, R. L.; Walgenbach, D. D.; Bohls, R. A.; spraying system and made cleanup significantly more Pooler, P. D.; Foster, R. N.; Reuter, K. C.; Swain, J. L. 1989. Imme- difficult. diate and 2nd-year effects of insecticide spray and bait treatments on populations of rangeland grasshoppers. Canadian Entomologist 121: Results from our study demonstrated that a low amount 589–602. of Dimilin active ingredient per acre with the 2F formula- Wang, T.; Fuller, B. W. 1991. Evaluation of Dimilin 1% bran bait for tion can be used in a large-scale control program manner grasshopper control on rangeland, 1990. Insecticide and Acaricide for control of grasshoppers on rangeland. Upon final Tests 16: 213. Environmental Protection Agency registration, Dimilin— because of its mode of action and its reduced spectrum of References Cited—Unpublished activity—could be an attractive option to be considered for controlling grasshoppers on rangeland. Jech, L. E.; Foster, R. N.; Colletto, D.; Walgenbach, D. D.; Bohls, R.; Rodriguez, G.; Burgess, J.; Meeks, W. K.; Queener, R. 1989. Field trials of chlorpyrifos and carbaryl bran baits for control of grasshop- References Cited pers on rangeland near Buffalo, South Dakota, 1988. In: Cooperative Grasshopper Integrated Pest Management Project, 1989 annual report. Bomar, R.; Lockwood, J. A. 1991. Evaluation of Dimilin bran-bait Boise, ID: U.S. Department of Agriculture, Animal and Plant Health for rangeland grasshopper control, 1990. Insecticide and Acaricide Inspection Service: 194–206. Tests 16: 212–213. Jech, L. E.; Foster, R. N.; Reuter, K. C.; Hirsch, D.; Cushing, W.; Jech, L. E.; Foster, R. N.; Colletto, D.; Walgenbach, D. D.; Roland, Walgenbach, D. D.; Bohls, R.; Herbaugh, L. L. 1992. Response of T. J.; Rodriguez, G. D.; Bohls, R.; Houston, R. D.; Meeks, W. K.; grasshopper species to carbaryl bran bait treatments. In: Cooperative Queener, R. L.; Jackson, C. L.; Dines, J. L.; Puclik, M. J.; Scott, Grasshopper Integrated Pest Management Project, 1992 annual report. A. K. 1993. Field evaluation of diflubenzuron and carbaryl bran baits Boise, ID: U.S. Department of Agriculture, Animal and Plant Health against grasshopper (Orthoptera: Acrididae) populations in South Inspection Service: 83–89. Dakota. Journal of Economic Entomology 86: 557–565.

VII.2–4 VII.3 Impact of Dimilin® on Nontarget Arthropods and Its Efficacy Against Rangeland Grasshoppers

Michael A. Catangui, Billy W. Fuller, and Arnie W. Walz

Introduction

Dimilin® (diflubenzuron) is a chitin-synthesis inhibitor and causes death in insects during the molting process (van Daalen et al. 1972, Post and Vincent 1973). Chitin, a nitrogenous polysaccharide, is the organic foundation of the exoskeleton of all insects and the entire phylum Arthropoda (Snodgrass 1935). Therefore, some concern exists that widespread use of diflubenzuron may affect not only the target insect pest but also nontarget arthropods that are essential for the diversity and stability of rangeland ecosystems. Other studies have shown the potential of diflubenzuron against rangeland grasshoppers (Foster et al. 1991 unpubl. and 1993 unpubl.). Figure VII.3–1—To minimize insecticide drift, spray booms are turned off well before this plane flies over a pond in the Great Plains. Our key research objective was to determine whether (Agricultural Research Service photo by John Kucharsky.) diflubenzuron negatively affected the abundance and diversity of nontarget arthropods (including ants, spiders, shrimp, water fleas, copepods, cladocerans, mayfly predatory beetles, and pollinator bees) in rangelands, and naiads, and midge larvae in treated (0.1 lb AI per acre) if so to determine if the effect was greater than the effect mosquito breeding areas in California. Adult aquatic of one of the current standard treatments. Another beetles, spiders, and mosquito fish were not affected by research objective was to develop additional data on the diflubenzuron even at the highest rates tested. Farlow et potential of diflubenzuron as an alternative insecticide al. (1978) studied the impact of diflubenzuron on nontar- against rangeland grasshoppers. get organisms of a Louisiana coastal marsh. Those authors reported significant reductions in amphipods, Previous studies indicate that diflubenzuron spares most dragonfly naiads, nymphs of corixid and notonectid bugs, nontarget arthropods. Ables et al. (1975) reported as well as adult hydrophilid beetles in marshlands treated diflubenzuron to be harmless to a pupal parasitoid of the six times with 0.025 lb AI per acre (28 g AI per ha) over house fly. Compared to dimethoate-treated poultry farms an 18-month period. On the other hand, significant in North Carolina, diflubenzuron-treated farms had increases were observed among mayfly naiads, larvae of greater parasitoid abundance and species diversity. In noterid and dytiscid beetles, adult corixid bugs, and mos- cotton fields, Keever et al. (1977) observed that arthro- quito fish. Numerous immature and adult insects were pod predators belonging to orders Hemiptera, Coleoptera, listed as unaffected by the diflubenzuron treatments. and Neuroptera were not affected by diflubenzuron when it was sprayed aerially at 0.12 lb active ingredient (AI) The environmental fate and degradation of diflubenzuron per acre (0.14 kg AI per hectare). Wilkinson et al. (1978) in a laboratory model ecosystem, a soil bacterium, sheep evaluated various rates and formulations of diflubenzuron liver microsomes, and ultraviolet light were investigated on adult and immature stages of selected parasitoids and by Metcalf et al. (1975). They found diflubenzuron to be predators found in cotton fields. The authors found test moderately persistent in organisms such as algae, snails, insects to be unaffected by diflubenzuron even at high caterpillars, and mosquito larvae but efficiently degraded concentrations except for immatures of a lacewing by mosquito fish, however. Ecological magnification species. may not be a problem: the lowest concentration of diflubenzuron was found in the mosquito fish, at the top In contrast, diflubenzuron may be detrimental to some of the model food chain. Sheep liver microsomes and the freshwater crustaceans and immature aquatic insects (fig. soil bacterium were not able to degrade diflubenzuron VII.3–1). Miura and Takahashi (1974, 1975) observed under the experimental conditions imposed. temporary population reductions in tadpole shrimp, clam

VII.3–1 Other studies have shown the potential of diflubenzuron We used rings to count live grasshoppers (fig. VII.3–2). against rangeland grasshoppers. Foster et al. (1991 Forty aluminum rings, each 0.1 m2, were arranged in unpubl.) reported aerial treatments of diflubenzuron spray grids near the center of each plot. We counted grasshop- at 0.015, 0.030, and 0.045 lb AI per acre to reduce pers within each ring using a tally counter. Sweep-net second- and third-instar grasshoppers as well as a stan- samples determined grasshopper species and their age dard treatment of carbaryl (0.5 lb AI per acre) after composition. 1 week. Foster’s team showed reductions for all treatments in the range of 94 to 96 percent after 2 weeks. Under simulated control program operational conditions, Foster et al. (1993 unpubl.) reported two formulations of diflubenzuron at 0.0156 AI per acre and a carbaryl standard performed equally well (control ranged from 87 to 91 percent). Our Study in South Dakota

Different rates and formulations of were tested in an open rangeland near Ludlow (Harding County), SD, during the 1993 season. Dimilin 2F (0.0075 and 0.015 lb AI/acre) and Dimilin 25W (0.015 lb AI/acre) were compared with Sevin® 4-Oil (0.5 lb AI/acre) and untreated Figure VII.3–2—A grasshopper’s eye view of the kind of ring field plots. The lower rate of Dimilin 2F was evaluated only crews use to delimit a sampling spot before counting resident ’hop- for efficacy against grasshoppers. The remaining treat- pers. (Agricultural Research Service photo by John Kucharsky.) ments were evaluated for impact on nontarget arthropods and efficacy against grasshoppers. We used a completely Sampling for nontarget arthropods was carried out before randomized design with each treatment replicated four and after treatment application. The malaise and pitfall times. A fixed-wing airplane applied chemical treat- traps were run a week before treatment, then resumed ments over 40-acre plots from July 2 to July 7, 1993. 1 week after the last chemical treatment application. Traps were maintained continuously thereafter, and Our study used pitfall traps to sample soil surface- catches were collected at weekly intervals for 10 weeks associated nontarget arthropods (ants, spiders, predatory from July to September. Plot and trap location markers beetles, and scavenger beetles). A pitfall trap consisted remained onsite over the winter months, and an addi- of a wide-mouth 1- qt canning jar filled with approxi- tional sample was collected about 1 year after treatment. mately 4 inches of mineral oil. Each pitfall trap was bur- We took grasshopper counts from rings and sweep-net ied so that the opening was flush with the soil surface. samples (fig. VII.3–3) once before chemical treatment The oil killed and temporarily preserved crawling insects and at weekly intervals for 7 weeks after treatment. that fell into the traps. Six pitfall traps spaced 15 ft apart Additional grasshopper counts and samples were taken and arranged in hexagonal pattern were installed near the the end of season (11 weeks after treatment). center of each 40-acre plot. We sorted nontarget arthropod samples and counted them Malaise traps were used to sample flying nontarget in the laboratory. Arthropods were identified to family arthropods such as parasitic and predatory wasps, lace- level then grouped according to their biological function wings, flies, and pollinator bees. Each malaise trap was a (such as predator, parasite, scavenger, or pollinator). 12- by 4- by 6-ft rectangular tent made of nylon screen Identification of ants to the species level (Wheeler and that intercepted and directed flying insects to killing jars. Wheeler 1963) was used to calculate a measure of species Two malaise traps were placed near the center of each diversity referred to as the probability of interspecific 40-acre plot. encounter (PIE) (Hurlbert 1971, Washington 1984).

VII.3–2 (3) grasshoppers. Each group was sampled using techniques appropriate for their mobility and biological characteristics.

Impact of Dimilin on Soil Surface-Associated Non- target Arthropods.—There were four major groups of soil surface-associated arthropods: (1) ants (order Hymenoptera: family Formicidae), (2) spiders (order Araneae: families Agelenidae, Amaurobiidae, Clubio- nidae, Dictynidae, Gnaphosidae, Hahniidae, Lycosidae, Mimetidae, Philodromidae, Salticidae, Tetragnathidae, Theridiidae, and Thomisidae), (3) predatory beetles (order Coleoptera: families Carabidae, Cicindelidae, Histeridae, Meloidae, and Staphylinidae), and (4) scavenger beetles (order Coleoptera: families Scarabaeidae, Silphidae, and Tenebrionidae).

In terms of biological function on the rangeland ecosystem, ants may be regarded as both general predators and scavengers (Wheeler and Wheeler 1963). All spiders are predators (Kaston 1972). Beetles belonging to families Carabidae (ground beetles), Cicindelidae (tiger beetles), Staphylinidae (rove beetles), and Histeridae (hister beetles) are also general predators (Borror and DeLong 1964). Blister beetle (Meloidae) larvae feed on grasshopper eggs, but adults are considered pests of certain crops. Scavengers were composed of families Scarabaeidae (scarab beetles), Silphidae (carrion beetles), and Tenebrionidae (darkling beetles). Certain scarabs like the dung beetle feed on cattle manure; carrion beetles feed on dead animal carcasses. Darkling beetles feed on decay- Figure VII.3–3—Sweep-netting grasshoppers is a labor-intensive but ing plant materials but some, like the false wireworms, time-tested method for sampling insect populations. (Agricultural feed on the roots of wheat and are considered pests. All Research Service photo by John Kucharsky.) arthropods mentioned above are important components of the rangeland food chain because they are potential food Hurlbert defined PIE as the probability that two individu- for vertebrate animals like birds, frogs, mice, moles, and als encountered at random in a community will belong to shrews. different species. In our present paper, PIE may be inter- preted as the probability that two individual ants ran- In general, Dimilin 2F (0.015 lb AI/acre), Dimilin 25W domly encountered in rangeland will be of different (0.015 lb AI/acre), and Sevin 4-Oil (0.5 lb AI/acre) did species. The higher the probability, the more diverse, not significantly reduce the number of ants, spiders, and presumably more stable, is the ant community. predatory beetles, or scavenger beetles from 7 to 76 days after treatment (DAT). Even at 1 year after treatment Findings and Discussion (350 to 357 DAT), no significant reductions in any of the soil surface-associated arthropods were detected. Ant Arthropods collected from the experimental site were numbers temporarily (49 to 55 DAT) declined after grouped arbitrarily as follows: (1) soil surface-associated Dimilin 2F and Sevin 4-Oil treatments by 43 and 56 per- nontarget arthropods, (2) flying nontarget arthropods, and cent, respectively. The temporary decline in ant numbers

VII.3–3 may or may not be due to chance alone. What is impor- dominant grasshopper species. Grasshopper age struc- tant is that ant numbers rebounded immediately and that ture was 46.8, 24.6, 23.5, 3.7, 0.2, and 0.1 percent for 1st, in most of the sampling periods, the Dimilin and Sevin 2d, 3d, 4th, and 5th instars and adults, respectively, at treatments were consistently shown to have no detrimen- 0 DAT. This age composition was ideal for a chitin- tal effects on ant numbers. Additionally, ant diversity synthesis inhibitor like Dimilin because the majority of (based on PIE calculations) was not significantly affected grasshoppers had several molts remaining in their life by the Dimilin or Sevin treatments from 7 to 357 DAT. cycle. This result may indicate that no ant species was particularly susceptible to the Dimilin and Sevin treatments at All Dimilin treatments were comparable to Sevin 4-Oil the dosages studied. starting at 14 DAT. From 14 DAT to 49 DAT, grasshopper numbers in the Dimilin- and Sevin-treated plots were Impact of Dimilin on Flying Nontarget Arthropods.— significantly lower than those of the untreated plots. The arthropods collected in malaise traps were sub- Dimilin provided consistent grasshopper control from divided into the following 3 groups: (1) pollinator bees 14 DAT to 49 DAT; Sevin-treated plots revealed tempo- (order Hymenoptera: families Apidae, Halictidae, rarily elevated grasshopper numbers at 35 DAT and Colletidae, Andrenidae, and Megachilidae), (2) predators 42 DAT. No differences between plots treated with (order Hymenoptera: families Sphecidae, Pompilidae, Dimilin at different rates or formulations were detected and Vespidae; order Diptera: families Asilidae and after 14 DAT. Therevidae; order Coleoptera: family Coccinelidae; order Neuroptera: families Chrysopidae, Hemerobiidae, and Dimilin was not as effective as Sevin at 7 DAT. This Myrmeleontidae), and (3) parasites (order Hymenoptera: delayed response is most likely due to its mode of action. families Ichneumonidae, Braconidae, Tiphiidae, Dimilin exerts its effect at molting while Sevin (a cholin- Chalcididae, Chrysididae, Mutillidae, Proctotrupidae, and esterase inhibitor) acts at any time of development. Pteromalidae; order Diptera: families Bombyliidae and Grasshopper population reductions (adjusted for natural Nemestrinidae). population changes) in Dimilin-treated plots ranged from 65 percent to 90 percent from 14 DAT to 49 DAT. In In general, no significant reductions in flying nontarget this study, all treatments lost effectiveness against grass- arthropods were observed in the Dimilin 2F, Dimilin hoppers by 76 DAT. For more information about 25W and Sevin 4-Oil treatments. Dimilin 25W reduced diflubenzuron efficacy on rangeland grasshoppers, see predator numbers during the 15- to 20-DAT period by chapter VII.2, “Dimilin Spray for Reducing Rangeland 59 percent. Predator numbers subsequently recovered, Grasshopper Populations.” and in most of the sampling periods, no significant reductions in predator numbers were observed. A temporary In summary, our study showed that Dimilin and Sevin decline of 18 percent in parasite numbers was recorded in sprays did not appear to significantly reduce the abun- the Dimilin 2F treatment at 35 to 41 DAT. No significant dance of soil-surface-associated or flying nontarget reductions were observed in the number of pollinator arthropods while providing good grasshopper control in bees. About 1 year after treatment (350 to 357 DAT), no rangeland. Our observations extended only through significant reductions in numbers of predators, parasites about 1 year after treatment. Interpretation of our results or pollinators were observed for any treatment. is limited to this period.

Efficacy of Dimilin Against Rangeland Grasshop- pers.—Nineteen grasshopper species were present on the 800-acre experimental area immediately before spraying (0 DAT). Melanoplus sanguinipes F., M. infantilis Scudder, and Trachyrhachys kiowa Thomas were the

VII.3–4 References Cited Snodgrass, R. E. 1935. Principles of insect morphology. New York: McGraw-Hill. 667 p. Ables, J. R.; West, R. P.; Shepard, M. 1975. Response of the house fly and its parasitoids to Dimilin (TH-6040). Journal of Economic Ento- van Daalen, J. J.; Meltzer, J.; Mulder, R.; Wellinga, K. 1972. Selec- mology 68: 622–624. tive insecticide with a novel mode of action. Naturwissenschaften 59: 312–236. Borror, D. J.; DeLong, D. M. 1964. Introduction to the study of insects. New York: Holt, Rinehart and Winston. 819 p. Washington, H. G. 1984. Diversity, biotic and similarity indices: a review with special relevance to aquatic ecosystems. Water Research Farlow, J. E.; Breaud, T. P.; Steelman, C. D.; Schilling, P. E. 1978. 18: 653–694. Effects of the insect growth regulator diflubenzuron on non-target aquatic populations in a Louisiana intermediate marsh. Environmen- Wheeler, W. P.; Wheeler, J. 1963. The ants of North Dakota. Grand tal Entomology 7: 199–204. Forks, ND: University of North Dakota Press. 326 p.

Hurlbert, S. H. 1971. The nonconcept of species diversity: a critique Wilkinson, J. D.; Biever, K. D.; Ignoffo, C. M.; Pons, W. J.; Morrison, and alternative parameters. Ecology 52: 577–586. R. K.; Seay, R. S. 1978. Evaluation of diflubenzuron formulations on selected insect parasitoids and predators. Journal of the Georgia Ento- Kaston, B. J. 1972. How to know the spiders, 3d ed. Dubuque, IA: mological Society 13: 226–236. Wm. C. Brown Co. 272 p. References Cited–Unpublished Keever, D. W.; Bradley, J. R.; Ganyard, M. C. 1977. Effects of diflubenzuron (Dimilin) on selected beneficial arthropods in cotton Foster, R. N.; Reuter, K. C.; Wang, T.; Fuller, B. W. 1991. Field fields. Environmental Entomology 6: 732–736. evaluations of diflubenzuron (Dimilin) spray for control of grasshoppers on rangeland. In: Cooperative Grasshopper Integrated Pest Man- Metcalf, R. L.; Lu. P.; Bowlus, S. 1975. Degradation and environmen- agement Project, 1991 annual report. Boise, ID: U.S. Department of tal fate of 1-(2,6-diflurobenzoyl)-3-(4-chlorophenyl) urea. Journal of Agriculture, Animal and Plant Health Inspection Service: 79–86. Agricultural Food Chemistry 23: 359–364. Foster, R. N.; Reuter, K. C.; Black, L.; Daye, G.; Vigil, E. L.; Walz, Miura, T.; Takahashi, R. M. 1974. Insect developmental inhibitors. A.; Radsick, B. 1993. Field evaluation of aerial application of two for- Effects of candidate mosquito control agents on nontarget aquatic mulations of diflubenzuron for rangeland grasshopper control under organisms. Environmental Entomology 3: 631–636. simulated control program operational conditions. In: Cooperative Grasshopper Integrated Pest Management Project, 1993 annual report. Miura, T.; Takahashi, R. M. 1975. Effects of the IGR, TH6040, on Boise, ID: U.S. Department of Agriculture, Animal and Plant Health nontarget organisms when utilized as a mosquito control agent. Inspection Service: 93–100. Mosquito News 35: 154–159.

Post, L. C.; Vincent, W. R. 1973. New insecticide inhibits chitin synthesis. Naturwissenschaften 60: 431–432.

VII.3–5

VII.4 An Attempt at Classical Biological Control of Rangeland Grasshoppers With Entomophaga grylli, Pathotype 3

Donald L. Hostetter

The primary objective of this project, conceived and Administrative policies and technical procedures within developed by R. I. Carruthers, was to develop and imple- Federal agencies and the State of North Dakota in effect ment a classical biological control program against at the time were observed and provided guidance for rangeland grasshoppers using an Australian isolate introduction. Permission was granted for field studies in (pathotype 3) of the grasshopper obligate Entomophaga North Dakota (Carruthers et al. 1989 unpubl., and in grylli (Zygomycetes: Entomophthorales) species complex press). (Ramoska et al. 1988). Pathotype 3 was isolated from Praxibulus sp. grasshoppers in Australia in 1985 by R. S. The goals of proposed releases were to reduce popula- Soper and R. J. Milner during an epizootic (grasshopper tions of economically important grasshoppers on western outbreak) (Milner 1985). rangeland to, or below, threshold densities; to establish pathotype 3 as a biorational agent that would augment The project was based on the collaborative findings of native fungi; and to determine the plausibility of future Soper and Milner and a 5-year study of the two native large scale releases throughout the Western United States species designated pathotype 1 and 2 in Arizona and New by PPQ’s Plant Protection Laboratories. Mexico (Carruthers and Humber 1988 unpubl.). Pathotype 3 was introduced into susceptible grasshopper Implementation of the attempt (Carruthers and Humber populations at several sites in McKenzie County in 1989, 1988 unpubl.) was through the U.S. Department of Agri- 1990, and 1991 and at two sites near Delta Junction, AK, culture, Animal and Plant Health Inspection Service, in 1990 (Carruthers et al. 1989 unpubl., 1990 unpubl., Plant Protection and Quarantine (USDA, APHIS, PPQ), 1991 unpubl.). Grasshopper Integrated Pest Management (GHIPM) Project initiated in 1987. Introduction was by randomly releasing laboratory- infected fifth-instar and adult M. differentialis (Thomas), The proposed and pursued approach was the introduction, each injected with 10 µl of 104 pathotype 3 protoplasts, establishment, evaluation, and dispersal of an Australian into grasshopper populations in alfalfa/mixed grass or isolate, pathotype 3, to augment the two native crested wheatgrass fields with no history of pathotype 1 pathotypes (1 and 2). Previous laboratory studies indi- or 2 fungus infection. Each field was about 44.5 acres cated that pathotype 3 had a wider host range than either (18 ha) in size. Releases were made at 2- to 3-day inter- of the native species plus other attributes that led to its vals 3 days postinjection (just prior to death of the grass- selection (Ramoska et al. 1988). hopper). Weekly releases in lots of 500 infected grasshoppers totalled from 500 to 3,500 at each site. These fungi, along with other biotic agents could theo- retically provide long-term, nonchemical suppression of The initial release of pathotype 3 was made July 24, grasshoppers between outbreaks. 1989. Five hundred (500) infected grasshoppers were released in an alfalfa/mixed-grass hayfield at Wold’s An ecological survey of sites with histories of grasshop- ranch (T153N, R97W, Sec. 33), 25 miles north of per populations and densities suitable for introduction Watford City, ND. Incidence of fungus infection among was made within the 17 Western States. The Little Mis- grasshoppers within this release site was 13 percent souri National Grasslands in McKenzie County, ND, was 2 weeks after the release (Carruthers et al. 1989 unpubl.). selected as the initial study area for field evaluation of pathotype 3 (Carruthers et al. 1989 unpubl.). Additional releases of ca. 500 per day were made at Wold’s on July 8, 11, 15, 19, 25, and 30, 1990. A 20- The use of biological control methods for grasshopper percent incidence of infection was observed at this site management, and specifically the introduction of the within 2 weeks of the 1990 releases. No additional Australian fungus, was supported by the membership of releases were made at this site after 1990. the McKenzie County Grazing Association, Watford City, ND.

VII.4–1 Similar releases were made in crested wheatgrass fields This project and plans for future releases of nonnative at three other sites in McKenzie County, ND, during pathogens and parasites within the GHIPM Project 1990. Incidence of fungus infection among grasshoppers caused intense debate among certain researchers and at these locations was less than 3 percent. Low incidence between agency administrators in 1991 (Bomar and of infection in these fields was attributed to the open Lockwood 1991, Lockwood 1993a and b, Carruthers and canopy of the crested wheatgrass, which likely resulted in Onsager 1993). In August 1991, amid the beginning con- a less favorable habitat for the fungus (Carruthers et al. troversy of the legality and wisdom of this approach, the 1990 unpubl.). Seasonal monitoring of grasshopper principal investigator (R. I. Carruthers) was reassigned, populations at these sites (1991–94) has failed to detect and the project was transferred from Ithaca, NY, to me at fungus-infected individuals. Kimberly, ID.

The releases of pathotype 3 into M. sanguinipes popula- Additional documentation was drafted and submitted tions at two sites in Alaska were considered unsuccessful (April 1992) and revised and resubmitted (October 1992) in that only a single sporulating cadaver was recovered 2 seeking a policy decision on the need for an environmen- weeks after release. Grasshopper populations at these tal assessment (EA) before proceeding with additional release sites have been monitored annually for incidence releases of pathotype 3. of fungal infection. Additional releases of pathotype 3 are stalemated. No Overwintering of pathotype 3 was thought to occur in releases of pathotype 3 have been made since June 1991. Wold’s field based on recovery of sporulating M. Efforts since that time have been relegated to monitoring bivittatus (Say) cadavers in June, 1991. Fungal mortality (population densities, composition, species fluctuations, among grasshoppers at this site reached 26 percent in incidence of mortality due to fungus infection, dispersal 1991 even though no additional introductions were made studies) in the release field and surrounding areas. (Carruthers et al. 1991 unpubl.). Laboratory studies were conducted to establish basic Releases of infected grasshoppers (500–1,000 each) were parameters of conidia production, germination and via- made on land managed by the U.S. Army Corps of Engi- bility, and dose/mortality curves, as well as mass inocula- neers near Lake Sakakawea (T154N, R95W, Sec. 32) on tion methods that would be required if the project was to June 6, 8, 11, and 13, 1991. The incidence of fungus be assumed and enlarged by PPQ’s Plant Protection infection at this location reached 25 percent 2 weeks after Laboratories. the last release. No additional releases were made after June 13, 1991. Grasshopper populations at this site con- The development of DNA probe technology for separa- tinued to be monitored for incidence of fungal disease tion and identification of three Entomophaga spp. of the through 1994. Populations and incidence of fungal infec- E. grylli complex has also been successful. Cooperation tion have been diminishing since 1991. between USDA’s Agricultural Research Service staff scientists at Ithaca, NY, and Kimberly, ID; the Boyce The initial success in North Dakota was encouraging, and Thompson Institute for Plant Science; and the University a plan for additional releases of 150,000 infected M. of Toronto, Scarborough campus led to the development differentialis (10,000 per week per location for 5 con- of a positive DNA identification probe whereby patho- secutive weeks) at 3 other locations was drafted. Addi- types 1, 2, and 3 can be separated and positively identi- tional releases were contingent upon production and fied (Bidochka et al. 1995). This is a critical accomplish- supply of suitable hosts by a commercial insectary in ment and provides a tool necessary to delineate dispersal Colorado. The number of sites and infected grasshoppers and distribution of pathotype 3 in the field. to be released was based on available human and fiscal resources as well as host population densities.

VII.4–2 References Cited Ramoska, W. A.; Hajek, A. E.; Ramos, M. E.; Soper, R. S. 1988. In- fection of grasshoppers (Orthoptera: Acrididae) by members of the Bidochka, M. J.; Walsh, S.R.A.; Ramos, M. E.; St. Leger, R. J.; Entomophaga grylli species complex (Zygomycetes: Entomop- Silver, J. C.; Roberts, D. W. 1995. Pathotypes in the Entomophaga thorales). Journal of Invertebrate Pathology 52: 309–313. grylli species complex of grasshopper pathogens differentiated by molecular methods. Journal of Applied and Environmental Micro- References Cited—Unpublished biology 61: 556–560. Carruthers, R. I.; Humber, R. A. 1988. Development of Entomophaga Bomar, C. R.; Lockwood, J. A., eds. 1991. Biological control of west- grylli as a biological control agent of grasshoppers. In: Cooperative ern rangeland grasshoppers with exotic predators and parasitoids: Grasshopper Integrated Pest Management Project, 1988 annual report. potential benefits, costs, and alternatives. Bull. B-958. Laramie, WY: Boise, ID: U.S. Department of Agriculture, Animal and Plant Health Wyoming Agricultural Experiment Station. Inspection Service: 330–335.

Carruthers, R. I.; Onsager, J. A. 1993. Perspective on the use of exotic Carruthers, R. I.; Humber, R. A.; Ramos, M. E. 1989. Development of natural enemies for biological control of pest grasshoppers (Orthop- Entomophaga grylli as a biological control agent of grasshoppers. In: tera: Acrididae). Environmental Entomology 22: 885–904. Cooperative Grasshopper Integrated Pest Management Project, 1989 annual report. Boise, ID: U.S. Department of Agriculture, Animal and Carruthers, R. I.; Ramos, M. E.; Larkin, T. S.; Hostetter, D. L.; Soper, Plant Health Inspection Service: 229–234. R. S. [In press.] The Entomophaga grylli (Fresenius) Batko species complex: its biology, ecology and use for biological control of pest Carruthers, R. I.; Humber, R. A.; Ramos, M. E. 1990. Development of grasshoppers. Memoirs of the Entomological Society of Canada. Entomophaga grylli as a biological control agent of grasshoppers. In: Cooperative Grasshopper Integrated Pest Management Project, 1990 Lockwood, J. A. 1993a. Environmental issues involved in biological annual report. Boise, ID: U.S. Department of Agriculture, Animal and control of rangeland grasshoppers (Orthoptera: Acrididae) with exotic Plant Health Inspection Service: 220–231. agents. Environmental Entomology 22: 503–518. Carruthers, R. I.; Humber, R. A.; Ramos, M. E. 1991. Development of Lockwood, J. A. 1993b. Benefits and costs of controlling rangeland Entomophaga grylli as a biological control agent of grasshoppers. In: grasshoppers (Orthoptera: Acrididae) with exotic organisms: search Cooperative Grasshopper Integrated Pest Management Project, 1991 for a null hypothesis and regulatory compromise. Environmental annual report. Boise, ID: U.S. Department of Agriculture, Animal and Entomology 22: 904–914. Plant Health Inspection Service: 185–189.

Milner, R. J. 1985. Field tests of a strain of Entomophaga grylli from the USA for biocontrol of the Australian wingless grasshopper, Phaulacriduim vittatum. In: Chapman, R. B., ed. Proceedings of the 4th Australian conference on grassland invertebrate ecology; 13–17 May 1985; Canterbury, UK. Christchurch, NZ: Caxton Press: [inclusive pagination unknown].

VII.4–3

VII.5 Lab Studies and Field Trials With the Fungus Beauveria bassiana Against Grasshoppers

R. Nelson Foster, K. Christian Reuter, Jim Britton, and Cliff Bradley

More than 150 years ago, the Hyphomycete fungus In 1987, Mycotech Corporation in Butte, MT, isolated, Beauveria bassiana was recognized as the cause of a dis- from an infected grasshopper found in Montana, a strain ease fatal to insects (Steinhaus 1967). B. bassiana is a of Beauveria bassiana that is virulent (disease-causing) common insect pathogen (an agent that causes disease) to several grasshopper species in laboratory bioassays. found on all continents except Antarctica (Humber 1992). Since that time, Mycotech has developed and refined pro- Hundreds of isolates of the fungus, including five from duction capabilities to the point that large-scale commer- grasshoppers, are listed in the U.S. Department of Agri- cialization is planned upon the final development of an culture (USDA) collection of Entomopathogenic Fungal acceptable formulation for delivering the pathogen to Cultures (Humber 1992). grasshoppers in the field. The following summarizes some of the research conducted since early 1991 in the In the fungus’ life cycle, conidia (spores) adhere to the development of formulations of Beauveria bassiana grasshopper cuticle (part of the exoskeleton). The usable against grasshoppers on rangeland. conidia germinate, and the germ tube penetrates the cuticle. The fungi replicate inside the insect haemocoel Laboratory Studies, 1991–93 (body cavity) in the form of blastospores (spores produced by a budding process). Degradative enzymes During this period, we conducted more than 20 different destroy the internal structures of the grasshopper. replicated studies. The objectives provided for (1) developing equipment and procedures for our laboratory stud- When in sufficient quantity, the fungus causes sickness ies, (2) studying the effect of Beauveria bassiana on within 3 days. The grasshopper reduces its feeding and different age groups of grasshoppers, (3) comparing of becomes immobile. Typically, infected grasshoppers die formulations, and (4) comparing the virulence of differ- between 4 and 10 days after infection depending on their ent batches of commercially produced B. bassiana. species, age, and size, and the dose of conidia. After death, under conditions of high humidity, blastospores Test formulations were sprayed from a tower apparatus in form hyphae (filaments of the vegetative structure of the the lab to simulate aerially applied sprays (fig. VII.5–2). fungus) that emerge through the insect’s cuticle, sporu- Applications were conducted according to a detailed late (produce spores), and cover the insect in a character- standard operating procedure (Foster and Reuter 1991 istic white growth (fig. VII.5–1). unpubl.). Laboratory-reared Melanoplus sanguinipes grasshoppers supplied by South Dakota State University were used for all studies. All tests focus on a dose of 1 × 1013 (1 trillion) spores/acre as a standard. Depending on the specific test protocol, we sprayed grasshoppers and/or live vegetation upon which the grasshoppers were to be confined.

When grasshoppers were sprayed, third instars through adult stages were sprayed singly or in groups consisting of from 5 to 20 grasshoppers per group. After spraying, the grasshoppers were monitored daily for death, usually for 2 weeks. In tests where grasshoppers were sprayed, fresh food was provided to surviving grasshoppers daily, and dead grasshoppers were held singly under high humidity conditions for observance of sporulation. Figure VII.5–1—An immature rangeland grasshopper, Melanoplus sanguinipes, exhibits the fungus Beauveria bassiana, which caused its death. (Photo by K. Christian Reuter.)

VII.5–1 In studies where untreated grasshoppers were confined on sprayed vegetation, we showed a significant decrease in mortality on vegetation that had been exposed to sunlight for longer than 24 hours (fig. VII.5–4). However, two formulations currently under development show promise for extending protection beyond 24 hours.

Third-, fourth-, and fifth-instar grasshoppers were easily infected and very susceptible to sprays equivalent to 1 × 1013 spores/gal/acre. However, compared to these results, two separate studies with adult grasshoppers showed a greatly reduced level of mortality at the same dose. Sub- sequent studies in which adults with amputated wings were sprayed showed that reduced mortality in adults cannot be attributed to physical protection provided by wings, which shield a major portion of the abdomen from the spray.

We conducted several studies to compare spores from different productions and to evaluate shelf life. Spores stored in oil for up to 1 year performed as well as dry conidia powder stored for an equal period. A 1992 spring production as well as a new isolate both performed similarly to spores produced in 1991. However, a 1992 fall production sampled resulted in some inconsistencies during the physical spraying. Slightly cooler temperatures during the spray operation may have affected the sprayability of the formulation. Also, a new harvesting method at the production facility resulted in some larger Figure VII.5–2—Spray tower used to simulate aerially applied sprays particles of spore powder, increasing spray problems. for bioassaying grasshoppers in the laboratory. (APHIS photo by Lonnie Black.) Field Studies—1991

Initial studies demonstrated the superiority of an oil for- A 9-acre rangeland plot near Edgemont, SD, infested mulation over a water formulation. A typical example of with predominantly second- and third-instar grasshoppers results from one of these tests is shown in figure VII.5–3. of mixed species, was aerially sprayed with an oil formu- In later studies where candidate field formulations were lation containing 8 × 1012 spores/gal/acre (fig. VII.5–5). compared, we focused primarily on different oil types Grasshopper moralities measured in this plot were com- with various additives selected for ultraviolet light pro- pared to a similar untreated adjacent plot (Foster et al. tection and emulsion stabilization (formulation stability). 1991 unpubl.). Two petroleum oils performed equally well as base carri- ers; however, one is significantly less expensive. We We evaluated mortality on six grasshopper species by found that formulations involving emulsifiable concen- collecting grasshoppers from both plots after application trates tend to be more difficult to spray consistently in the and confining them in (1) small rearing cups (fig. VII.5– laboratory. However, our results indicate that such com- 6), which we moved to the laboratory for daily monitor- pounds may provide higher mortality in field application. ing, and (2) bottomless field cages (fig. VII.5–7) estab-

VII.5–2 lished after treatment in both plots. Additionally, 0.1-m2 In field plots, counts of unconfined populations in treated rings (Onsager and Henry 1977) were used to delimit and untreated plots showed average differences in mortal- counting areas for estimating total field populations of ity that ranged from about 39 percent to 63 percent at 3 to grasshoppers. 15 days after treatment (fig. VII.5–8).

Beauveria bassiana caused mortality in all six species of We also used field cages to determine the general manner the grasshoppers tested. Both grasshoppers held in rear- in which grasshoppers pick up the spores. Immediately ing cups in the laboratory and those caged on native vege- after application, grasshoppers from the untreated plots tation in the field demonstrated significant mortality in were collected and caged in the treated area to determine treated populations compared to untreated populations. pickup through feeding activity. Treated grasshoppers Some species were killed faster than others, but we do were caged in the untreated plot to determine the mortal- not know if this is due to inherent susceptibility or behav- ity associated with direct contact. Treated grasshoppers ioral differences between the species. were caged in the treated plot to determine the total mortality, and untreated grasshoppers were caged in the In rearing cups, the average reduction of all species com- untreated plot as a control. bined in treated populations was about 96 percent at 8 days after treatment. Mortality in the controls during At 11 days after treatment, there were no significant dif- the same period was about 34 percent. In field cages, the ferences in grasshopper mortality between the direct mean reduction of all species combined was 79 percent deposition, feeding activity, or combined direct deposi- and 11 percent for treated and untreated populations, tion/feeding activity treatments. All three treatments respectively, at 9 or 10 days after treatment. showed significantly greater mortality than the untreated

Mean percent mortality (10 replications) 100

Bb+oil/1 gal Bb+oil/1/2 gal 80 Bb+H2O/1 gal Oil carrier Untreated

0 0 2 4681012 14 Days after treatment

Figure VII.5–3—Mortality of caged grasshoppers treated with experimental formulations of Beauveria bassiana at 1 × 1013 conidia per acre.

VII.5–3 Mean percent mortality 100

Bb/Oil-24% petroleum jelly 80 Bb/Oil-12% petroleum jelly Bb/H2O-19% petroleum jelly Bb/H2O-12% petroleum jelly Bb/Std Oil 60 Untreated

0 0 h 24 h 48 h 72 h Hours of sunlight exposure to sprayed and unsprayed vegetation

Figure VII.5–4—Effect of grass treated with selected formulations of Beauveria bassiana and exposed to several periods of sunlight on grasshopper survival after 9 days. All treatments were applied at a volume of 1 gal/acre containing 1 × 1013 spores.

Figure VII.5–5—The first aerial application of the fungus Beauveria Figure VII.5–6—Four-ounce rearing cups used to confine test bassiana was applied at 1 gal/acre to a rangeland plot near Edgemont, grasshoppers after they have been treated. (APHIS photo by SD in 1991. (Photo by Cliff Bradley.) R. Nelson Foster.)

VII.5–4 check. Our data indicate that pickup may occur through intervals. These samples were washed, diluted, and either direct impingement (direct striking by spray drop- placed on selective agar plates, where fungus colonies let) or feeding activity. We do not know if the feeding developed from each colony-forming unit. The colonies activity component is simply due to contact with the then were counted to estimate the number of viable (liv- mouthparts of the grasshopper during feeding or actual ing) conidia. ingestion of spores. Untreated grasshoppers exposed to the treated vegetation We evaluated the short-term residual activity of the in the field approximately 10 hours after application died spores by caging untreated grasshoppers approximately at about 3.3 times the mortality rate of untreated grass- 10 hours after treatment in the treated plot. Survival of hoppers over the same period of time, 11 days. The the conidia on vegetation was evaluated in the sprayed delayed exposure demonstrates the infectivity of spores plot by taking vegetation samples at three posttreatment at least 10 hours after field application and indicates that, in field situations, at least several hours are available for a grasshopper to become infected with the fungus. Results of the study to determine survival of conidia on vegetation in the field showed relatively uniform cover- age in the plot and indicated no loss of activity over at least the first 10 hours after application. Field Studies—1992

Three adjoining 9-acre rangeland plots near Amidon, ND, infested with predominately fourth- and fifth-instar grasshoppers of mixed species were the basis for studies in 1992. One plot was aerially sprayed with 9.5 × 1012 spores/64 oz/acre. One plot was sprayed with 64 oz/acre of the oil carrier (without spores), and the other plot was Figure VII.5–7—Bottomless field cages used to confine test left untreated for comparison (Foster et al. 1992 unpubl.). grasshoppers in the field are inspected carefully to determine the daily insect mortality. (APHIS photo by R. Nelson Foster.) Mortality evaluations were conducted as in 1991, by confining, after treatment, the six predominant grasshopper species in cages held in the laboratory or in the field. The methods used for maintaining the cages and confirm- ing fungus-induced death by sporulation were similar to those employed in 1991. Reduction in the total field population was again estimated by using 0.1-m2 rings to delimit counting areas.

In this study, the aerial application of B. bassiana resulted in substantial mortality of all six species of grasshoppers evaluated. Both grasshoppers held in rearing cups in the laboratory and those caged on native vegetation in the field demonstrated significant mortality in fungus-treated populations compared to untreated populations and populations treated with oil only. These Figure VII.5–8—Mortality of unconfined field populations of grass- results were generally similar to those obtained in 1991, hoppers is estimated by counting grasshoppers in metal rings. and again time to mortality varied among species, begin- (APHIS photo by R. Nelson Foster.)

VII.5–5 ning in as little as 3 days for some species and as much as rangeland plots located near Amidon, ND, infested with 4 to 6 days for other species. These differences may be predominantly second-, third-, and fourth-instar stages of attributed to individual species susceptibility or a result grasshoppers of mixed species. Two formulations of of behavioral avoidance, which limits physical exposure Beauveria bassiana spores were each applied to eight of individual species to direct impingement of the spray plots. One treatment consisted of 9.9 × 1012 spores/64 oz/ droplet. acre in an oil formulation, and the other treatment consisted of 9.4 × 1012 spores/64 oz/acre in an oil plus addi- In rearing cages, the mean reduction of all species com- tive (adjuvant) formulation. An oil-only treatment was bined in treated populations was 95 percent at 8 days applied at 64 oz/acre to four plots. Carbaryl was sprayed after treatment. During the same time period, mortality at 20 oz/acre (0.5 lb/active ingredient [AI] per acre) to in the untreated population and the population treated four plots as a standard treatment for comparison. Four only with oil was 10 percent and 4 percent, respectively. plots were left untreated to determine the natural changes Three species common to both the 1991 and 1992 studies in the grasshopper population and for comparison with all demonstrated very similar responses to the aerially applied treatments. applied B. bassiana treatment. In field populations, estimates were again made using In field cages, the mean reduction for 5 of the 6 species 0.1-m2 rings. A monitoring site located near the center of confined in treated populations was 91 percent at 15 to 17 each 40-acre plot consisted of 40 rings arranged in a days following treatment. This reduction compared to circle with rings separated by 5 paces. Field cages were mortality during the same period in the untreated popula- placed adjacent to the ring site in each plot after the treat- tion and the population treated only with oil of 23 percent ment was sprayed. Sprayed grasshoppers of two of the and 11 percent, respectively. The sixth species in the dominant species were confined in these cages in a man- study was reduced much quicker: 100-percent mortality ner similar to that employed in 1991 and 1992 field occurred by the eleventh day. Its counterparts in the studies. untreated plots and the plots treated with oil showed 26 percent and 16 percent reduction during the same Additional field cages were set up in each fungus- and period. oil-only treated plot and in the untreated plots. These cages were used to study the residual activity of Comparisons of the in-field posttreatment population Beauveria bassiana over a 5-day period after treatment. estimates in single, small plots are difficult to interpret. Untreated grasshoppers were confined in some cages on High densities of grasshoppers, sparse vegetation, small the day of treatment and on each of the 5 days following plot size, and local movement all contribute to confound- treatment. ing estimates of nonrestricted in-field populations. Com- pared to 1991, in-field mortality was lower in this study. Unfortunately, the study’s value was lessened by measur- In 1992, apparent mortality at 9 days after treatment was able rain (heavy at times) that occurred on 9 of the 13 only about 20 percent. We did note that vegetation in the days that population estimates were made. During the 1992 study was much sparser than in the 1991 study and entire study, measurable rain was recorded on 15 of 21 may have offered the spores less protection from sun- days. light. Using large field plots in future studies should reduce many of the difficulties commonly encountered Although incomplete, analysis of counts from rings to when comparisons of in-field grasshopper populations on date shows that the carbaryl standard was statistically rangeland are attempted. superior to all other treatments at each of the posttreatment interval readings. Good performance of carbaryl Field Studies—1993 under these conditions was expected and is consistent with two of our previous studies where carbaryl was used We focused studies for the first time in 1993 on larger (Foster et al. 1991 unpubl. and Foster et al. 1993 unpubl.). plots than previously used (Foster et al. 1993 unpubl.). All other experimental treatments (including the That year, we aerially sprayed 24 adjoining 40-acre untreated checks) showed erratic results, undoubtedly

VII.5–6 confounded by the weather conditions experienced small-plot field trials in the west African countries of during the study, and were statistically inseparable. Cape Verde and Mali. Fungal spores were applied at a rate of 1 × 1013 per acre. Low-volume application of an Results from the field cages for the two species studied at oil-based formulation (27 ounces to 2 quarts per acre) 15 days after treatment indicated that both fungus treat- was made with hand-held spinning disc sprayers. High- ments and the carbaryl treatment produced mortality sig- volume application of an emulsifiable formulation (2–10 nificantly greater than what occurred in the untreated gal/acre) was made with motorized or hand-pumped populations. However, mortality in the field cages was backpack sprayers. Spores were also formulated on somewhat lower than in 1991 and 1992 for the one spe- wheat bran bait with a molasses sticker. cies that was common to studies in all 3 years. In all trials, 80 to 100 percent of treated, caged insects Residual activity was evident only during the day of died from Beauveria bassiana infection after 7 days. treatment. Beyond 1 day, no significant differences in More significantly, replicated 5-acre blocks in Cape mortality were detected between fungus-treated or Verde, treated with either oil-formulated or emulsion- untreated grasshoppers. formulated fungus, showed approximately 50 percent population density reductions measured in the field after Under the conditions of this study, evaluations of 7 days. It is quite encouraging that the insect population unproven formulations are confounding and inconclusive in these tests consisted primarily of older nymphs and at best. However, there is no doubt that carbaryl per- adults, which have demonstrated more resistance to the formed well under these conditions and that the current fungus in laboratory bioassays. formulation of Beauveria bassiana will need to be improved if it is to be employed under these conditions, Mycotech and Montana State University have taken part or excluded from use under such conditions. Additional in an expedition to Madagascar to collect new fungal replicated studies to obtain information on the original pathogens of locusts and grasshoppers. The fungi iso- objectives of the 1993 field study and new formulation lated from infected insects are presently being examined evaluations are planned for the future. for virulence, target specificity, production characteristics, and impact on mammals. The government of Mada- Summary of Additional Foreign Studies gascar is particularly interested in using fungi to treat locust populations before the insects expand out of their During the past 5 years, Mycotech has been working to recessionary (nonoutbreak) areas. When a suitable fun- develop fungal pathogens of locusts and grasshoppers for gus is identified, field trials will begin. use in integrated pest management (IPM) programs in Africa. This work is in collaboration with Montana State These promising results indicate that fungal insecticides University, the U.S. Agency for International Develop- may be able to play an important role in grasshopper/ ment, and several African government agencies. These locust control. This field experience in the harsh African efforts were undertaken to devise alternatives to chemical conditions will continue to yield information valuable to grasshopper/locust control measures commonly used in the development of fungal insecticides for North Africa. Fungi can fit well into an IPM scheme because America. they provide control alternatives where chemical insecticides are inappropriate. In fact, because of their rela- Summary and Conclusion tively slow action, fungi will work best as part of a continuous pest-control strategy, where they can be A strain of the entomopathogenic fungus Beauveria applied before populations are able to reach damaging bassiana has been isolated from U.S. grasshoppers by levels. Mycotech Corporation. Development of mass production capabilities with a potential for large-scale commerciali- A Mycotech strain of the fungus Beauveria bassiana has zation has resulted in extensive testing of the commer- been tested against grasshoppers and locusts in several cially produced fungus for use against grasshoppers and

VII.5–7 locusts. Laboratory studies have demonstrated the insec- References Cited—Unpublished ticidal value of the fungus against several species of grasshoppers and locusts. In 1991, 1992, and 1993, we Foster, R. N.; Reuter, K. C. 1991. Using a spray tower to simulate conducted field studies using cages to demonstrate suc- aerially applied liquid sprays for bioassaying insects in the laboratory: standard operating procedure. Phoenix, AZ: U.S. Department of Agri- cessful control of several species of confined grasshop- culture, Animal and Plant Health Inspection Service, Methods Devel- pers in the United States when liquid formulations of opment, Rangeland Section. 5 p. Beauveria bassiana were aerially applied with conventional commercial application equipment. Results of Foster, R. N.; Reuter, K. C.; Black, L. R.; Vigil, E.; Daye, G.; Walz, field studies with unconfined grasshoppers in this country A.; Radsick, B. 1993. Field evaluation of aerial applications of two are inconclusive to date. Foreign field studies on uncon- formulations of diflubenzuron for rangeland grasshopper control under simulated control program operational conditions. In: Coopera- fined populations showed good potential for providing tive Grasshopper Integrated Pest Management Project, 1993 annual control. Results from the last 3 years suggest the poten- report. Boise, ID: U.S. Department of Agriculture, Animal and Plant tial for controlling several species of grasshoppers and Health Inspection Service: 93–100. locusts using a liquid formulation of B. bassiana, as a bioinsecticide, and applied with conventional aerial Foster, R. N., Reuter, K. C.; Bradley, C.; Britton, J.; Black, L.; Drake, S.; Leisner, L.; Tompkins, M.; Fuller, B.; Hildreth, M.; Kahler, E.; application equipment. Radsick, B. 1992. Development of the fungus Beauveria bassiana as a bio-insecticide for grasshoppers on rangeland. In: Cooperative Grass- References Cited hopper Integrated Pest Management Project, 1992 annual report. Boise, ID: U.S. Department of Agriculture, Animal and Plant Health Humber, Richard A. 1992. Collection of entomopathogenic fungal Inspection Service: 207–215. cultures. ARS-110. Washington, DC: U.S. Department of Agriculture, Agricultural Research Service. 177 p. Foster, R. N.; Reuter, K. C.; Bradley, C. A.; Wood, P. P. 1991. Pre- liminary investigations on the effect of Beauveria bassiana on several Onsager, J. A.; Henry, J. H. 1977. A method for estimating the density species of rangeland grasshoppers. In: Cooperative Grasshopper Inte- of rangeland grasshoppers (Orthoptera: Acrididae) in experimental grated Pest Management Project, 1991 annual report. Boise, ID: U.S. plots. Acrida 6: 231–237. Department of Agriculture, Animal and Plant Health Inspection Ser- vice: 203–210. Steinhaus, E. A. 1967. Insect microbiology. New York and London: Hafner Publishing Co. 763 p. Foster, R. N.; Reuter, K. C.; Tompkins, M. T.; Bradley, C.; Britton, J.; Underwood, N.; Black, L. R.; Vigil, E.; Daye, G.; Radsick, B.; Fuller, B.; Brinkman, M.; Kahler, E. 1992. Development of the fungus Beauveria bassiana as a bio-insecticide for grasshoppers on rangeland. In: Cooperative Grasshopper Integrated Pest Management Project, 1993 annual report. Boise, ID: U.S. Department of Agricul- ture, Animal and Plant Health Inspection Service: 233–238.

Foster, R. N.; Reuter, K. C.; Wang, T.; Fuller, B. W. 1991. Field evaluation of diflubenzuron (Dimilin) spray for control of grasshoppers on rangeland. In: Cooperative Grasshopper Integrated Pest Man- agement Project, 1991 annual report. Boise, ID: U.S. Department of Agriculture, Animal and Plant Health Inspection Service: 79–86.

VII.5–8 VII.6 Beauveria bassiana for Mormon Crickets

D. A. Streett and S. A. Woods

Introduction (2) Germination of the conidium and penetration of the insect cuticle by a germ tube from the conidium, The first crops planted by the Mormon settlers in Utah were damaged by the insect now referred to by the com- (3) Growth of the fungus inside the insect body (hemo- mon name “Mormon cricket” (Cowan 1990). The Mor- coel) and eventual death of the insect, mon cricket, Anabrus simplex Haldeman, is not a cricket at all but a longhorned grasshopper from the family Tetti- (4) Penetration of the fungus to the surface of the dead goniidae (fig. VII.6–1). This pest can reach outbreak lev- insect and formation of conidia (plural of conidium) els before Mormon crickets begin migrating into range under conditions of high relative humidity, and and cropland. Mormon crickets can cause significant damage when bands of huge numbers of insects move (5) Dispersal of the conidia to locations where they may onto cropland in the Western United States (Pfadt 1991, encounter susceptible insects and start the process MacVean 1990, Swain 1944). Our studies evaluated the again. effectiveness of a fungal pathogen, Beauveria bassiana, to suppress Mormon cricket populations. Among the insect-pathogenic fungi that follow this pattern of development is Beauveria bassiana. It is commonly known as the white-muscardine fungus because of the characteristic white covering of conidia (spores) found on the surface of dead insects. Insect cadavers infected with the fungus are transformed into white, mummified bodies resembling in appearance a bonbon candy (“muscardin” means “bonbon” in French [Steinhaus 1949]). Isolate of B. bassiana for Mormon Cricket

The B. bassiana strain used in these studies was origi- nally obtained from Mycotech Corporation in Butte, MT. Mycotech has obtained Environmental Protection Figure VII.6–1— The Mormon cricket is mainly a pest on rangelands Agency registration of this Beauveria strain for the sup- but sometimes moves into planted crops and causes economic dam- pression of several insect pests, including grasshoppers age. (Agricultural Research Service file photo K4797–1.) and Mormon crickets. Mycotech recently developed a solid culture system for the production of B. bassiana How Beauveria bassiana Works conidia (Goettel and Roberts 1992). Mycotech prepared and supplied a B. bassiana dry conidia powder for the laboratory studies and B. bassiana formulated in oil (OF) Interest in insect–fungi interactions has centered, for the and in an emulsible suspension (ES) for the 1992 and most part, on the pathogenic (disease-causing) nature of 1993 Idaho field trials (Onsager et al. 1992, Kemp and fungi and their use as microbial control agents. Unlike Streett 1993). other insect pathogens that must be eaten to infect insects, fungi can infect an insect through its cuticle (outer skin). The development of fungi pathogenic to Laboratory Studies insects typically follows this pattern: Conidia were suspended in ES1 and ES2 oil and applied µ (1) Attachment of an infectious stage (called a conidium to Mormon crickets as 0.08 L (microliter) droplets or spore) to the insect cuticle, beneath the pronotum (on the thorax) at dosages ranging from 0 to 106 spores per Mormon cricket. Mormon crickets were reared individually in plastic cups and main-

VII.6–1 tained in an incubator at 77 °F (25 °C). Mormon crickets Four replicates of 200 adult Mormon crickets each were were fed every 2 days with romaine lettuce, kale, and treated with 5 × 105 or 5 × 106 conidia in oil according to wheat bran. Mortality was recorded during feeding, and the procedures described by Kemp and Streett, 1993. A a damp cotton ball was added to cups containing cadav- check preparation consisting of oil without conidia and ers. The cadavers were then stored at room temperature an untreated control were included for each replicate. for 4–6 days to diagnose Beauveria infection by observ- Each treatment within a replicate was separated into two ing the characteristic white muscardine appearance on the groups and reared either individually in an incubator at insect surface. 77 °F or transferred to field enclosures. Four field enclosures 16 ft2 (1.5 m2) for each treatment were stocked with

The median lethal dose (LD50) is commonly used to 25 Mormon crickets. Mormon crickets were fed lettuce assess the infectivity of a pathogen. The LD50 for the B. daily. Counts of Mormon crickets were made for each bassiana isolate against fifth-instar Mormon crickets at cage, and cadavers were collected for incubation in cups 12 days was 1,000 conidia (fig. VII.6–2). The two oil with a moistened cotton ball to diagnose Beauveria infec- formulations that were compared in laboratory assays tion (Kemp and Streett 1993). showed no consistent differences in overall mortality or percentage of Mormon crickets with confirmed infections (table VII.6–1).

Mortality (%) 100

1,000 × 103 100 × 103 80 10 × 103 Oil control Control

0 0 2 4681012 14 Days after inoculation

Figure VII.6–2—Cumulative mortality among fifth-instar Mormon crickets in a bioassay of Beauveria bassiana.

VII.6–2 Table VII.6–1—Laboratory comparison of ES1 versus ES2 oil as a carrier for Beauveria bassiana. Cumulative mortality and incidence of infection for Mormon crickets.

Mortality Infection

Dose 1ES1 ES2 ES1 ES2

Conidia/ grasshopper Percent

0 3446 812

102 50 38 20 18

103 71 87 42 44

104 90 98 65 62

1 ES = emulsifiable suspension.

Adult Mormon crickets that were inoculated with 5 × 106 Treatments were replicated four times, and treatments conidia per Mormon cricket showed a significant differ- within each replicate were applied on the same day ence in mortality in laboratory versus field cages (fig. (weather permitting) in the sequence outlined by Onsager VII.6–3). Adult Mormon crickets reared in the field et al. (1992). An ultralow-volume sprayer (North Ameri- enclosures survived more than 3 weeks longer than can Micron) was used for the applications. After applica- Mormon crickets reared in the laboratory. One possible tion, Mormon crickets were collected from each arena for explanation for these results is that Mormon crickets in rearing. Approximately 30–50 Mormon crickets per the field use a behavioral thermoregulation to increase arena were reared individually in the laboratory; mortal- body temperature to a point that restricts fungal develop- ity and infection data were recorded as described earlier. ment and allows the insect to survive. Three field cages (16 ft2/cage) were each stocked with 30–50 Mormon crickets from each arena and covered Field Studies with chicken wire to keep out birds. Mormon crickets were fed lettuce and sagebrush daily. Mormon crickets Field trials against Mormon crickets were conducted near were counted daily, and cadavers were collected and St. Anthony, ID. Oil (ES1 oil) and clay–oil–water incubated in cups with a moistened cotton ball to diag- (COW)—100 g clay: 1 liter (L) oil: 2 L water)—formula- nose Beauveria infection. tions were applied at rates of 4.9 ( 1011 and 4.9 × 1012 conidia/acre (1.2 × 1012 and 1.2 × 1013 conidia per ha) and Results differed somewhat between the formulations that application volumes of 0.9 and 2.7 qt/acre (2.5 and 7.5 L/ were used in the field. The statistical results suggested ha). Each replicate consisted of 10 arenas of 14.4 yd2 (12 that the ES1 formulation produced less mortality but m2) constructed of aluminum flashing approximately 10– similar rates of infection than the OF formulations at the 18 inches (25–45 cm) in height. Each arena was stocked 2.7 qt/acre application volume. There were no differ- with more than 250 Mormon crickets prior to application. ences in overall mortality or infection rates between the

VII.6–3 0.9 qt/acre and 2.7 qt/acre application volumes of oil Conclusions alone formulations. It should be noted that while the differences in mortality between formulations at the 2.7 qt/ A detailed understanding of the disease dynamics of the acre application volume may have been statistically sig- B. bassiana isolate will be necessary before this product nificant, they were not substantial (80 v. 74 percent at the can be considered for use in an integrated pest manage- low conidia concentration). ment program. Gaining this understanding will entail both laboratory and field studies to evaluate short-term The application rate of conidia had a more substantial and longrange impacts of Beauveria on Mormon crickets. impact on both the overall mortality and percentage of The effects of cannibalism, behavioral fever, and host confirmed infections. Adjusted for controls, overall mor- behavior will need further evaluation before the potential tality averaged 55 percent and 89 percent for the low and of B. bassiana as a microbial control agent against high conidia concentrations, respectively. All compari- Mormon crickets can be determined. Formulation of sons between conidia concentrations were statistically B. bassiana for Mormon cricket control will also require significant. additional research.

Mortality (%) 100

1,000 × 103 Lab 100 × 103 Lab 80 1,000 × 103 Field 100 × 103 Field

0 0 5 10 15 20 2530 35 Days after inoculation

Figure VII.6–3—Cumulative mortality among adult Mormon crickets treated with Beauveria bassiana in the lab and reared in the lab or in field cages.

VII.6–4 References Cited Steinhaus, E. A. 1949. Principles of insect pathology. New York: McGraw Hill. 757 p. Cowan, F. T. 1990. The Mormon cricket story. Spec. Rep. #31. Bozeman, MT: Montana State University and Montana Agricultural Swain, R. B. 1944. Nature and extent of Mormon cricket damage to Experiment Station. 42 p. crop and range plants. Tech. Bull. 866. Washington, DC. U.S. Depart- ment of Agriculture. 44 p. Goettel, M. S.; Roberts, D. W. 1992. Mass production, formulation and field application of entomopathogenic fungi. In: Lomer, C. J.; References Cited—Unpublished Prior, C., eds. Biological control of locusts and grasshoppers. Wallingford, UK: CAB International: 230–238. Onsager, J. A.; Streett, D. A.; Woods, S. A. 1992. Grasshopper pathogen evaluation. Cooperative Grasshopper Integrated Pest Management MacVean, C. 1990. Mormon crickets: a brighter side. Rangelands Project, 1992 annual report. Boise, ID: U.S. Department of Agricul- 12: 234–235. ture, Animal and Plant Health Inspection Service: 179–186.

Pfadt, R. E. 1991. Mormon cricket. In: Field guide to common west- Kemp, W. P.; Streett, D. A. 1993. Grasshopper pathogen evaluation. ern grasshoppers. Sta. Bull. 912. Laramie, WY: U.S. Department of Cooperative Grasshopper Integrated Pest Management Project. 1993 Agriculture, Animal and Plant Health Inspection Service and Wyo- annual report. Boise, ID: U.S. Department of Agriculture, Animal and ming Agricultural Experiment Station. 4 p. Plant Health Inspection Service: 217–224.

VII.6–5

VII.7 Effects of the Fungus Beauveria bassiana on Nontarget Arthropods

Mark A. Brinkman, Billy W. Fuller, and Michael B. Hildreth

Introduction Spray-tower laboratory bioassays as developed by Foster and Reuter (1991) also were used at SDSU to determine Beauveria bassiana is currently being developed as a the effects of B. bassiana on nontarget insects. A spray potential bioinsecticide alternative to traditional chemical tower consist of a small airbrush, such as artists use, pesticides for controlling grasshopper populations. Cur- mounted on a stand and connected to an air pump. A rently, Nosema locustae is the only other nonchemical solution of fungal conidia (sporelike stage) can then be treatment registered for control of grasshoppers on range- injected into the airstream and sprayed onto the insects. land. B. bassiana offers at least two major advantages This method of conidia application should more closely over N. locustae: (1) B. bassiana appears to kill grass- simulate the field aerial application of conidia than would hoppers more rapidly than does N. locustae (see VII.5 applying the conidia in a large single drop or by sub- and I.3), and (2) Beauveria does not rely on the ingestion merging the insects in a solution of conidia (Foster and of its spores in a bait formulation by grasshoppers but is Reuter 1991). capable of directly penetrating through their exoskeleton (Goettel 1992). Adult yellow mealworm beetles (Tenebrio molitor) were evaluated with the bioassay because they are easily Unfortunately, B. bassiana may possess at least one acquired commercially and have therefore served as potential disadvantage. Unlike the narrow specificity of research models in many laboratory studies. The species N. locustae for orthopterans (i.e., grasshoppers, locusts T. molitor belongs to the family Tenebrionidae, which is and crickets), B. bassiana is known to infect a wide vari- an important group of beetles on western rangeland. This ety of insects (Goettel 1992). The wide specificity of beetle was selected also to represent the many species of Beauveria is of concern because distribution of its beetles evaluated in the field study whose population conidia into the environment also might diminish benefi- levels appeared unaffected by the release of B. bassiana cial insect populations. Attempts have been made to conidia into their locality. select strains of B. bassiana with increased specificity for grasshoppers by selecting stains isolated from grasshop- According to Goerzen et al. (1990), alfalfa leafcutting pers (Prior 1992). bees (Megachile rotundata) should be considered in evaluations of potential microbial agents. Unfortunately, Mycotech Corporation (Butte, MT) has mass-produced a the low numbers of alfalfa leafcutting bees recovered in strain of B. bassiana isolated from an infected grasshop- field plots prior to the North Dakota study made it impos- per found in Montana. Laboratory and field studies have sible to evaluate the effects of B. bassiana on this spe- indicated that this strain is infectious and lethal in con- cies. Therefore, M. rotundata was evaluated in the fined populations of several species of grasshoppers (see laboratory bioassay. Spray tower bioassays were first VII.5). However, no information existed on its virulence conducted with fourth-instar Melanoplus sanguinipes in nontarget insects. grasshoppers in order to standardize our results with those reported in VII.5. In 1993, South Dakota State University (SDSU) assisted the Animal and Plant Health Inspection Service (APHIS) Field Studies by monitoring the population levels of nontarget arthropods in a B. bassiana field study located near Methods.—Thirteen days prior to aerial treatments, sam- Amidon, ND (Brinkman 1995). The grasshopper control pling traps were placed in 4 control plots, 4 carbaryl data for this study are described in chapter VII.5. Impor- plots, and 4 plots that were to receive B. bassiana at the tant nontarget arthropods on rangeland include beneficial rate of 9.9 trillion spores/64 oz/acre in oil formulation. pollinators (flies and bees), predators (spiders, ants, Ground-dwelling arthropods were sampled with the use ground beetles, robber flies, green lacewings, brown of pitfall traps. Pitfall traps are widemouth quart canning lacewings, antlions, ladybird beetles, blister beetles, and jars placed in the ground with the opening level with the wasps), parasites or parasitoids (flies and several hymen- soil surface. Ground-dwelling arthropods were captured, opterans) and general scavengers (ants and darkling killed, and preserved as they fell into the jars, which con- beetles). tained 70 percent alcohol.

VII.7–1 Aerial insects were sampled using malaise traps. Insects Laboratory Studies were captured by malaise traps as they flew into the netting, and instinctively crawled or flew up into jars at the Methods.—Fungal conidia (spores) and an oil carrier top. Sampling traps were left in plots for 5 days, and solution were supplied by Mycotech Corp. Aerial appli- then jars and samples were retrieved. Immediately after cation of B. bassiana was simulated in the laboratory treatments, jars were replaced in plots and retrieved every with the use of a spray tower. A favorable spray pattern 6 days for the duration of the summer season. Arthropod was established in practice tests with the oil solution and samples were taken to SDSU to be sorted and identified. the aid of oil-sensitive paper. Procedures, equipment and B. bassiana dosages were similar to those described in Results.—During the study period, an abnormally high VII.5 and were selected based on recommendations by level of precipitation fell on the study plots. The result- Foster and Reuter (1991). ing high moisture level was favorable for the natural outbreak of Beauveria infections identified in the control A total of 360 individuals of each species were tested in grasshoppers from the untreated plots. This natural the laboratory experiments. Prior to each spray event, Beauveria outbreak may then have been at least partially clean newsprint was placed on the floor of the spray responsible for the unexpected erratic results seen in this room. In addition, test insects (in groups of 10) were study in both the treated and untreated plots. slowed by cooling to 35 °F (1.7 °C). Thirty individuals were sprayed with air for approximately 15 seconds first Ant and spider abundance declined in all plots following and were kept as controls. Thirty insects were sprayed treatment but rebounded the next week. The sporadic with 0.09 mL of the oil carrier. Thirty insects were heavy precipitation that occurred following treatment sprayed with 0.09 mL of oil containing 2.64 billion may have resulted in decreased activity of those ground- conidia/mL. Treatments were replicated four times. dwelling arthropods, and thus diminished their chances of Insects were then observed for 10 days after treatment. falling in the pitfall traps. Therefore, the temporary decrease in ant and spider abundance did not appear to be Results.—Grasshoppers treated with B. bassiana began due to B. bassiana or carbaryl treatments. Ground beetle expiring on day 5. After 10 days, more than 73 percent (Carabidae) densities remained stable throughout the of treated grasshoppers had died. Mortality of beetles summer season. treated with B. bassiana was extremely low, and beetles did not appear to be susceptible to infection. Flies (Diptera) were the most prevalent aerial insects captured in malaise traps. Abundance of flying Diptera, B. bassiana was extremely virulent to alfalfa leafcutting Hymenoptera, Lepidoptera, Neuroptera, and Coleoptera bees. Alfalfa leafcutting bees sprayed with B. bassiana increased in all plots following treatments. B. bassiana began expiring on day 4. After 10 days, more than 87 and carbaryl applications did not result in any noticeable percent of alfalfa leafcutting bees had died. However, declines in aerial insect abundance. mortality of alfalfa leafcutting bees sprayed with oil and air (control) was low. Dead alfalfa leafcutting bees were Alfalfa leafcutting bees were very rare at the study site. individually placed in glass vials with a moist cotton ball Only three individual Megachilidae were collected in and were observed for evidence of infection. After malaise traps during the sampling season. The study site approximately 7 days, external sporulation of hyphae was dominated by mixed grasses, so there was little (filaments of the vegetative structure of the fungus) was attraction for pollinating bees. Consequently, we were observed on 99 percent of alfalfa leafcutting bees treated not able to determine if field applications of B. bassiana with B. bassiana. affected alfalfa leafcutting bees.

VII.7–2 Conclusions References Cited

Treatment of the study sites with B. bassiana caused no Brinkman, M. A. 1995. Non-target arthropod impacts and rangeland measurable permanent decrease in populations of any of grasshopper management using Beauveria bassiana (Bals.) Vuillemin. Doctoral dissertation. Brookings, SD: South Dakota State University. the monitored beneficial insects. This lack of effect 62 p. occurred during a time period when moisture levels in the fields were abnormally high, and thus, environmental Foster, R. N.; Reuter, K. C. 1991. Using a spraytower to simulate conditions should have been very good for the spread of aerially applied liquid sprays for bioassaying insects in the laboratory: the infection into beneficial insects. In fact, even some of standard operating procedure. Phoenix, AZ: U.S. Department of Agriculture, Animal and Plant Health Inspection Service, Methods the grasshoppers recovered from the control sites also Development, Rangeland Section. 5 p. were infected with Beauveria, but at low levels and most likely from a natural outbreak. Gaugler, R.; Costa, S. D.; Lashomb, J. 1989. Stability and efficacy of Beauveria bassiana soil inoculations. Environmental Entomology Spray-tower results on lab-reared grasshoppers were 21(6): 1239–1247. similar to those described in VII.5. The nonsusceptibility Goerzen, D. W.; Erlandson, M. A.; Moore, K. C. 1990. Effect of two of the Tenebrio molitor to B. bassiana in the spray-tower insect viruses and two entomopathogenic fungi on larval and pupal bioassay was consistent with Beauveria’s apparent lack development in the alfalfa leafcutting bee, Megachile rotundata (Fab.) of effect on beetles in the field study. The effects of B. (Hymenoptera: Megachilidae). The Canadian Entomologist 122: bassiana on alfalfa leafcutting bees were evaluated only 1039–1040. with the spray-tower bioassay because few bees were Goettel, M. S. 1992. Fungal agents for biocontrol. In: Lomer, C. J.; recovered in the field. Existing bioassay data indicate Prior, C., eds. Biological control of locusts and grasshoppers. that these insects are very susceptible to this strain of B. Wallingford, UK: C.A.B. International: 122–130. bassiana. Injury to the entire population of alfalfa leafcutting bees might be reduced through management. Prior, C. 1992. Discovery and characterization of fungal pathogens for locust and grasshopper control. In: Lomer, C. J.; Prior, C., eds. Bio- B. bassiana conidia can persist if protected from environ- logical control of locusts and grasshoppers. Wallingford, UK: C.A.B. International: 159–180. mental extremes (soil is the natural reservoir for conidia), but become nonviable after only a few hours of exposure to sunlight (Gaugler et al. 1989, see VII.5). Alfalfa leafcutting bees readily accept artificial nesting structures, which could be moved during spray operations and returned later.

VII.7–3

VII.8 Grasshopper Viruses

D. A. Streett and S. A. Woods

Introduction EPV Laboratory Studies

Insect poxviruses or “entomopoxviruses” (EPV’s) infect Cross-infection studies have been reported for only seven insects from the following five insect orders: Coleoptera grasshopper and locust EPV’s (Henry et al. 1985, Oma (beetles), Lepidoptera (moths and butterflies), Orthoptera and Henry 1986, Streett et al. 1990, Lange and Streett (grasshoppers and crickets), Diptera (flies), and Hymen- 1993). Relative susceptibility of grasshoppers to a given optera (bees and wasps). The grasshopper EPV’s are EPV is usually limited to grasshoppers within the same found in the genus Entomopoxvirus B, which also subfamily (Lange and Streett 1993). However, it is inter- includes viruses from Lepidoptera and Orthoptera esting to note that some grasshopper EPV’s have been (Esposito 1991). All grasshopper viruses are physically found to infect grasshoppers from several different sub- similar and have roughly the same deoxyribonucleic acid families. (DNA) size. They differ from EPV’s in other insect orders and other animal poxviruses. Indeed, there is no Henry and Jutila (1966) isolated the first grasshopper evidence to suggest any close relationship or similarity EPV from the lesser migratory grasshopper, Melanoplus between grasshopper entomopoxviruses and other viruses sanguinipes, a frequent pest on crops and rangeland. The of vertebrate or invertebrates (Langridge 1984). virus, referred to as the Melanoplus sanguinipes entomopoxvirus (MsEPV), infects mostly species in the Virus particles are embedded in a crystalline proteina- genus Melanoplus (Oma and Henry 1986). Grasshoppers ceous matrix referred to as an occlusion body (OB). infected with a sufficient amount of the virus develop OB’s vary in size from 3 to 12 microns (µm) in diameter slowly, are sluggish, and die from the effects of the virus and may each contain up to several hundred virus par- (Henry and Jutila 1966). ticles. Twelve µm equal about 1/20,000th of an inch. OB’s offer the virus particles some protection from envi- MsEPV is the only grasshopper EPV that has been grown ronmental conditions and are thought to be responsible in vitro (outside the body) (Kurtti et al. 1990 unpubl). for transmission of a virus from one grasshopper to The M. sanguinipes cell culture lines designated another. When OB’s are ingested by a grasshopper, the UMMSE–1A, UMMSE–4, and UMMSE–8 have proven virus particles are released and penetrate through the susceptible to infection by MsEPV. The UMMSE–4 cell digestive tract into the body of the grasshopper. Infection cultures show cytopathic effects (undergo cell changes) by grasshopper EPV’s appears to be restricted to the fat when inoculated with MsEPV. The virus produced in body, a tissue which is used to store food reserves and vitro is both infectious and virulent (poisonous) against metabolize food. After the virus particles enter a fat body M. sanguinipes. Occlusion bodies produced in vitro, cell, they replicate and pack the cytoplasm with new though, were somewhat smaller—each about 6 µm in OB’s that contain virus particles. Virus particles will diameter (1/40,000 of an inch)—than occlusion bodies also spread to other fat body cells until nearly all the cells produced in vivo (inside the body). The latter were each in the fat body are infected with virus (Henry et al. 1969, about 12 µm in diameter. Granados 1981). In the laboratory, mortality from MsEPV occurs in two EPV’s are the only viruses containing DNA that have distinct timeframes over 5 or more weeks. Infectious been found in field grasshoppers. Typically, an EPV will OB’s are not present in grasshoppers that die during the be named after the host species of the original isolation. first interval of mortality, so these cadavers are of little Following this convention, there are at least 15 grasshop- importance for pathogen transmission. As dosage per EPV’s reported in the literature (Henry and Jutila increases, the proportion of inoculated grasshoppers that 1966, Langridge et al. 1983, Oma and Henry 1986, die prior to OB formation increases dramatically. Conse- Henry et al. 1985, Wang 1994). quently, the proportion of infected grasshoppers that survive long enough to produce OB’s actually decreases

VII.8–1 with dosage (Woods et al. 1992). These observations Under high density conditions, there may be considerable suggest that the strategy for using this virus in an inte- competition for these cadavers with the larger individuals grated pest management program may well depend on the successfully defending the resource against smaller specific objectives at the time of application. Maximum intruding grasshoppers (O’Neill et al. 1993). When both transmission rates are likely to be attained by applying infected and uninfected cadavers were placed in the field, the virus at low rates, and so an EPV treatment may be an there were no significant differences in the number of appropriate strategy for grasshopper populations that are cadavers that were partially consumed (K. M. O’Neill, increasing in density. A high-density population that is unpublished data). already causing significant damage should be treated with high rates to cause substantial early mortality. EPV Field Studies

Sublethal effects that have been observed for virus- The Environmental Protection Agency granted an Experi- infected grasshoppers include a delay in development, mental Use Permit (EUP) for field evaluations of MsEPV reduction in food consumption, and potential reduction in in 1988. Field evaluations were conducted from 1988 to egg production by the female. All of these sublethal 1990. Human and domestic-animal safety studies were factors can have a profound effect on grasshopper completed, and no evidence of infectivity was detected in populations. any of the studies. Toxicology data to identify hazards that MsEPV might present to nontarget organisms were The delay in development was reported first by Henry et also conducted with no evidence of toxicity or pathoge- al. (1969) and later by Olfert and Erlandson (1991). In nicity (poisonous or disease-related effects) observed in some cases, grasshopper nymphs infected with MsEPV any of the animals examined in these studies. In addi- will remain 9 to 18 days longer in an instar. Total food tion, Vandenberg et al. (1990) did not observe reductions consumption by grasshoppers infected with MsEPV was in longevity or pathological effects when MsEPV was reduced by 25 percent at 5 days after infection and up to tested against newly emerged adult workers of the honey- 50 percent at 25 days after infection. This reduction in bee, Apis mellifera. food consumption in MsEPV-infected nymphs was directly related to dose. Field evaluations of the potential for using MsEPV for grasshopper control were conducted during 1989. Plots The effects of MsEPV infection on M. sanguinipes egg were treated with virus that was formulated in starch production are unclear. While it has been difficult to granules (McGuire et al. 1991). At 13 days after applica- thoroughly describe the effects of MsEPV on M. tion, prevalence (the number of diseased insects at any sanguinipes egg production, we have observed that de- given time) was estimated at 14 percent and 23 percent in velopment to the adult stage is delayed by infection, and the plots receiving the low or high application rates, none of the infected adults in our laboratory studies have respectively. Prevalence was estimated at 9.2 percent in produced any eggs. the control plots at 13 days after application, indicating that considerable dispersal between plots had already Routes of Transmission occurred (Streett and Woods 1990 unpubl.). Our field studies from 1989 emphasize the problems associated One of the more likely routes of EPV transmission is with evaluation of microbial insecticides against insects through the consumption of infected cadavers. Grasshop- with considerable dispersal capabilities. That we can pers will commonly consume other grasshoppers that are infect at least 23 percent of the population with a rate of sick or dying. When grasshopper cadavers were placed 10 billion OB’s/acre (24.7 billion OB’s/ha) is clear. The in the field, nearly 92 percent of the cadavers were almost actual infection levels, in view of the dispersal problem entirely consumed after 30 minutes (O’Neill et al. 1994). and early mortality from the pathogen, are probably much higher.

VII.8–2 References Cited Oma, E. A.; Henry, J. E. 1986. Host relationships of entomopoxviruses isolated from grasshoppers. In: Grasshopper symposium Esposito, J. J. 1991. Poxviridae. In: Francki, R.I.B.; Fraquet, C. M.; proceedings; March 1986; Bismarck, ND. Fargo, ND: North Dakota Knudson, D. L.; Brown, F., eds. Classification and nomenclature of Extension Service, North Dakota State University: 48–49. viruses. Archives of Virology, Supplementum 2: 91–102. O’Neill, K. M.; Streett, D.; O’Neill, R. P. 1994. Scavenging behavior Granados, R. R. 1981. Entomopoxvirus infections in insects. In: of grasshoppers (Orthoptera: Acrididae): feeding and thermal Davidson, E. W., ed. Pathogenesis of invertebrate microbial diseases. responses to newly available resources. Environmental Entomology [Place of publication unknown]: Allanheld, Osman: 101–126. 23: 1260–1268.

Henry, J. E.; Jutila, J. W. 1966. The isolation of a polyhedrosis virus O’Neill, K. M.; Woods, S. A.; Streett, D. A.; O’Neill, R. P. 1993. from a grasshopper. Journal of Invertebrate Pathology 8: 417–418. Aggressive interactions and feeding success of scavenging rangeland grasshoppers (Orthoptera: Acrididae). Environmental Entomology 22: Henry, J. E.; Nelson, B. P.; Jutila, J. W. 1969. Pathology and develop- 751–758. ment of the grasshopper inclusion body virus in Melanoplus sanguinipes. Journal of Virology 3: 605–610. Streett, D. A.; Oma, E. A.; Henry, J. E. 1990. Cross infection of three grasshopper species with the Melanoplus sanguinipes Henry, J. E.; Wilson, M. C.; Oma, E. A.; Fowler, J. L. 1985. Patho- entomopoxvirus. Journal of Invertebrate Pathology 56: 419–421. genic micro-organisms isolated from West African grasshoppers (Orthoptera: Acrididae). Tropical Pest Management 31: 192–195. Vandenberg, J. D.; Streett, D. A.; Herbert Jr., E. W. 1990. Safety of grasshopper entomopoxviruses for caged adult honey bees (Hymen- Lange, C. E.; Streett, D. A. 1993. Susceptibility of Argentine optera: Apidae). Journal of Economic Entomology 83: 755–759. melanoplines (Orthoptera: Acrididae) to entomopoxviruses (Entomopoxvirinae) from North American and African grasshoppers. Wang, L. Y. 1994. Surveys of entomopoxviruses of rangeland grass- Canadian Entomologist 125: 1127–1129. hoppers in China. Scientia Agricultura Sinica 27: 60–63.

Langridge, W.H.R. 1984. Detection of DNA base sequence homology Woods, S. A.; Streett, D. A.; Henry, J. E. 1992. Temporal patterns of between entomopoxviruses isolated from Lepidoptera and Orthoptera. mortality from an entomopoxvirus and strategies for control of the Journal of Invertebrate Pathology 43: 41–46. migratory grasshopper (Melanoplus sanguinipes (F.). Journal of Inver- tebrate Pathology 60: 33–39. Langridge, W.H.R.; Oma, E. A.; Henry, J. E. 1983. Characterization of the DNA and structural proteins of entomopoxviruses from References Cited—Unpublished Melanoplus sanguinipes, Arphia conspirsa, and Phoetaliotes nebrascensis (Orthoptera). Journal of Invertebrate Pathology 42: Kurtti, T. J.; Munderloh, U. G.; Ross, S. E.; Ahlstrand, G. G.; Streett, 327–333. D. A. 1990. Cell culture systems for production of host dependent grasshopper pathogens. In: Cooperative Grasshopper Integrated Pest McGuire, M. R.; Streett, D. A.; Shasha, B. S. 1991. Evaluation of Management Project, 1990 annual report. Boise, ID: U.S. Department starch-encapsulation for formulation of grasshopper (Orthoptera: of Agriculture, Animal and Plant Health Inspection Service: 246–251. Acrididae) entomopoxviruses. Journal of Economic Entomology 84: 1652–1656. Streett, D. A.; Woods, S.A. 1990. Grasshopper pathogen field evaluation: virus. In: Cooperative Grasshopper Integrated Pest Management Olfert, O. O.; Erlandson, M. A. 1991. Wheat foliage consumption by Project, 1990 annual report. Boise, ID: U.S. Department of Agricul- grasshoppers (Orthoptera: Acrididae) infected with Melanoplus ture, Animal and Plant Health Inspection Service: 210–217. sanguinipes entomopoxvirus. Environmental Entomology 20: 1720–1724.

VII.8–3

VII.9 Use of an Australian Parasite of Grasshopper Eggs as a Biological Control Agent

Richard J. Dysart

Introduction tions against grasshoppers certainly have not been exhausted, particularly with scelionid egg parasites. In order to increase the existing mortality level of any Worldwide in distribution, the species of the genus Scelio pest grasshopper, entomologists are generally limited to are all egg parasites of acridid grasshoppers and there are two biological control approaches: augmentation or no host records from any other group of insects (Great- introduction. In the former, some parasite or predator head 1963, Muesebeck 1972, Galloway and Austin 1984). species must be reared in great numbers and distributed evenly over the crop or rangeland to be protected. The Rationale for Classical Introduction.—Although there augmentation process must be repeated year after year as are several native Scelio spp. present in western North needed. In the introduction approach, a parasite or preda- America, they cause only minor levels of egg mortality. tor species, from outside of the system, is imported and The most abundant and most widespread of our native colonized, with the intention of obtaining permanent egg parasites is Scelio opacus (Provancher). During an establishment of the natural enemy. Ideally, the natural 8-year study in Wyoming, Lavigne and Pfadt (1966) enemy species would be colonized only once and would found only trace numbers of Scelio parasites in rangeland spread and distribute itself once established. grasshopper eggs. Results of a long-term study in Saskatchewan (Mukerji 1987) showed that egg parasitism Augmentative Approach by Scelio averaged about 5 percent and had no detectable impact on field populations. In my own field studies in In my opinion, using insect parasites or predators northeastern Montana and northwestern North Dakota augmentatively, as substitutes for chemical insecticides, from 1988 to 1994, egg-pod parasitism by native Scelio is not feasible for the control of grasshoppers. The chief spp. averaged 10.7 percent (Dysart 1995), but parasitism obstacle to this approach is the cost. Although certain of individual eggs was only 4.1 percent (Dysart 1994 Scelio egg parasites can be reared easily in the laboratory, unpubl.). the rearing process is dependent upon a constant supply of live grasshopper eggs of a certain age. Considering Although the ecological niche is occupied by several the immense areas that would require treatment with native parasites, their total impact on the eggs of pest parasites, plus the logistics of rearing and delivery, it is grasshoppers probably does not affect infestations. certain that the costs of using Scelio wasps Therefore, in 1989, I proposed to the Animal and Plant augmentatively would be unacceptable. Health Inspection Service (APHIS) that I try to import and establish an additional species of Scelio. If this new Classical Introduction Approach parasite became established on one or more of the destructive grasshoppers in the West, it could increase Historical.—According to a worldwide review article by egg mortality and thereby reduce initial densities of Prior and Greathead (1989), classical biological control nymphs. That scenario could greatly enhance the prob- of a grasshopper with scelionid wasps has been attempted ability of other indigenous (native) natural enemies main- on only one occasion. The attempt was made in Hawaii, taining suppression of pest grasshopper densities at or during 1930 and 1931, against the Chinese grasshopper, below economic thresholds for greater time intervals. Oxya chinensis (Thunberg), using two parasite species from Malaysia, Scelio serdangensis Timberlake and S. Periodic outbreaks probably would not be eliminated, but pembertoni Timberlake (Pemberton 1933, Clausen 1978). the interval between them might be lengthened or the Scelio serdangensis failed to establish, but S. pembertoni duration of outbreaks might be shortened. Introduction became established and is reported to have successfully of exotic parasites to help control indigenous pests is controlled the pest (Pemberton 1948, Clausen 1978). As controversial, but as pointed out by Huffaker et al. pointed out by various authors (Commonwealth Institute (1971), there is no pest that should be judged in advance of Biological Control 1981, Siddiqui et al. 1986, as not amenable to biological control. A good review Greathead 1992), the possibilities for classical introduc- article on this subject is presented by Carl (1982).

VII.9–1 Search for a Candidate Scelio in Australia.—In Sep- obtained emergence of adults of S. parvicornis from tember 1990 and again in 1992, my Australian colleagues 33 species, and it failed to emerge from egg-pods of and I collected egg-pods of several different grasshoppers 16 grasshopper species (Dysart 1993 unpubl.). About and locusts at 10 localities in the States of New South half of the 33 successful lab hosts of S. parvicornis are Wales, South Australia, and Western Australia. In Sep- considered to be our most serious rangeland pests (Hewitt tember 1992, we made collections in 11 different locali- 1977) (see also chapter VI.6). ties in the same states. A summary of these collections is found in Dysart (1993 unpubl.) and in Baker et al. (in Plans for Field Releases and Recovery Attempts.— press). In 1990, overall parasitism of egg-pods by Scelio Assuming that permission to release parasites was spp. was 28 percent (128 of 460 egg-pods), but was high- granted by the Federal and State authorities, I had est (36 percent) in Western Australia (66 of 181 egg- planned to proceed as follows: colonies of several thou- pods). During 1990, Scelio parvicornis Dodd was the sand adult parasites would be released over a period of most abundant parasite of the five species reared, and at several weeks at one or more sites in Arizona, Montana, one locality, Nungarin (Kittyea ranch), in Western Aus- and North Dakota. Prior to releases at proposed sites, tralia, it parasitized about 25 percent of the host egg-pods screened cages would be erected on sandy soil and fur- (Australian plague locust, Chortoicetes terminifera nished with wild female grasshoppers (M. sanguinipes). [Walker]). Two articles, Baker and Pigott (1993) and After egg-laying was well under way, adult parasites Baker et al. (in press), provide additional parasitism and would be introduced into the cages. The cages would be host-range information on S. parvicornis. The egg-pod removed the following spring, and during the next two parasitism figures from Australia are considerably higher seasons, egg-pods would be excavated at the site and held than those reported above for western North America. for emergence in the laboratory to determine if the Aus- tralian parasite had successfully overwintered. If Scelio Quarantine Screening in the United States.—Grass- parvicornis is released and becomes established, it will hopper egg-pods collected in Australia were kept chilled be necessary to conduct additional field studies to assess and were hand-carried to the Montana State University its impact on pest grasshopper populations. quarantine facility in Bozeman. There the eggs were allowed to hatch, and all Australian grasshopper nymphs Addendum.—I made my initial request to U.S. Depart- were identified and then destroyed. Of the five species of ment of Agriculture, APHIS, Plant Protection and Quar- Scelio that emerged from the 1990 collections, we inves- antine, Biological and Taxonomic Support (USDA, tigators selected Scelio parvicornis (Nungarin strain) as APHIS, PPQ, BATS) for permission to release Scelio our primary candidate, based on its dominant position in parvicornis in the summer of 1991. Periodically during the Australian collections and its ease of rearing in the 1992 and 1993, I provided BATS with revisions and sup- quarantine laboratory. port documents as they continued to prepare their risk assessment (Lakin 1994 unpubl.). The question of Rearing and Host-Range Tests.—Using nondiapausing whether or not the Australian parasite should be released eggs of a native pest grasshopper, Melanoplus in North America has been the subject of active debate in sanguinipes (Fabricius), as hosts, my research team was the literature, between Lockwood (1993a and b) and able to propagate a nondiapausing culture of S. Carruthers and Onsager (1993). Lockwood is opposed to parvicornis in the laboratory. Under our lab conditions, the field release of the parasite because he feels that its we produced a new generation of parasites about every potential host range is too broad, and he speculates that it 32 days. In laboratory comparison tests with the native S. might have a detrimental effect on benign, nonpest grass- opacus, females of the Australian S. parvicornis were hoppers as well as a few grasshoppers thought to be ben- clearly superior: they parasitized more egg-pods and eficial because they feed on rangeland weeds. Carruthers killed more eggs during their respective lifetimes (Dysart and Onsager believe that the release of the Australian egg 1991 unpubl.). In laboratory host-range tests, we parasite is warranted and that the risk of harm to nontar- exposed the Australian parasite to about 1,808 egg-pods get species is negligible at best. of 49 species of North American grasshoppers. We

VII.9–2 On April 6, 1994, I received word from the permitting Greathead, D. J. 1992. Natural enemies of tropical locusts and grass- agency, USDA, APHIS, PPQ, BATS, that my application hoppers: their impact and potential as biological control agents. In: for the release of Scelio parvicornis had been denied. As Lomer, C. J.; Prior, C., eds. Biological control of locusts and grasshoppers. Wallingford, UK: C.A.B. International: 105–121. a result, I have destroyed the laboratory colony and have abandoned my plans for field releases of the parasite. I Hewitt, G. B. 1977. Review of forage losses caused by rangeland still believe that the overall benefits of the proposed bio- grasshoppers. Misc. Publ. 1348. Washington, DC: U.S. Department of logical control introduction would outweigh any potential Agriculture, Agricultural Research Service: 1–24. risks, but for the time being, the outcome will remain a Huffaker, C. B.; Messenger, P. S.; DeBach, P. 1971. The natural matter of conjecture. enemy component in natural control and the theory of biological control. In: Huffaker, C. B., ed. Biological control. New York: Plenum References Cited Press: 16–67.

Baker, G. L.; Pigott, R. G. 1993. Parasitism of Chortoicetes termini- Lavigne, R. J.; Pfadt, R. E. 1966. Parasites and predators of Wyoming fera (Walker) (Orthoptera: Acrididae) eggs by Scelio parvicornis rangeland grasshoppers. Sci. Monogr. 3. Laramie, WY: Wyoming Dodd (Hymenoptera: Scelionidae). Journal of the Australian Entomo- Agricultural Experiment Station: 1–31. logical Society 32: 121–126. Lockwood, J. A. 1993a. Environmental issues involved in the biologi- Baker, G. L.; Dysart, R. J.; Pigott, R. G. In press. Parasitism of grass- cal control of rangeland grasshoppers with exotic agents. Environmen- hopper and locust eggs (Orthoptera: Acrididae) by Scelio spp. tal Entomology 22: 503–518. (Hymenoptera: Scelionidae) in southern Australia. Australian Journal of Zoology. Lockwood, J. A. 1993b. Benefits and costs of controlling rangeland grasshoppers (Orthoptera: Acrididae) with exotic organisms: search Carl, K. P. 1982. Biological control of native pests by introduced for a null hypothesis and regulatory compromise. Environmental natural enemies. Biocontrol News and Information 3(3): 191–200. Entomology 22: 904–914.

Carruthers, R. I.; Onsager, J. A. 1993. Perspective on the use of exotic Muesebeck, C.F.W. 1972. Nearctic species of Scelionidae (Hymen- natural enemies for biological control of pest grasshoppers (Ortho- optera: Proctotrupoidea) that parasitize the eggs of grasshoppers. ptera: Acrididae). Environmental Entomology 22: 885–903. Smithsonian Contribution to Zoology 122: 1–33.

Clausen, C. P. 1978. Orthoptera, Acrididae. In: Clausen, C. P., ed. Mukerji, M. K. 1987. Parasitism by Scelio calopteni Riley (Hymeno- Introduced parasites and predators of arthropod pests and weeds: a ptera: Scelionidae) in eggs of the two dominant melanopline species world review. Agric. Handb. 480. Washington, DC: U.S. Department (Orthoptera: Acrididae) in Saskatchewan. Canadian Entomologist of Agriculture, Agricultural Research Service: 9–10. 119: 147–151.

Commonwealth Institute of Biological Control. 1981. Possibilities for Pemberton, C. E. 1933. Introduction to Hawaii of Malayan parasites biological control of grasshoppers (Acridoidea) status paper. Plant (Scelionidae) of the Chinese grasshopper Oxya chinensis (Thun.) with Protection Bulletin 33(3&4): 157–160. life history notes. Proceedings of the Hawaiian Entomological Society 8(2): 253–264. Dysart, R. J. 1995. New host records for North American Scelio (Hymenoptera: Scelionidae), parasitic on grasshopper eggs (Ortho- Pemberton, C. E. 1948. History of the entomology department experi- ptera: Acrididae). Journal of the Kansas Entomological Society 68(1): ment station, H.S.P.A. 1904–1945. Hawaiian Planters’ Record 52: 74–79. 53–90.

Galloway, I. D.; Austin, A. D. 1984. Revision of the Scelioninae Prior, C.; Greathead, D. J. 1989. Biological control of locusts: the (Hymenoptera: Scelionidae) in Australia. Australian Journal of Zool- potential for the exploitation of pathogens. Food and Agriculture Plant ogy. Supplementary Series 99: 1–138. Protection Bulletin 37: 37–48.

Greathead, D. J. 1963. A review of the insect enemies of the Siddiqui, R. K.; Irshad, M.; Mohyuddin, A. I. 1986. Digest: Scelio Acridoidea (Orthoptera). Transactions of the Royal Entomological spp. as biocontrol agents of acridids. Biocontrol News and Informa- Society of London 114: 437–517. tion 7(2): 69–76.

VII.9–3 References Cited–Unpublished

Dysart, R. J. 1991. Impact of insect parasites and predators on grasshopper populations II. In: Cooperative Grasshopper Integrated Pest Management Project, 1991 annual report. Boise, ID: U.S. Department of Agriculture, Animal and Plant Health Inspection Service: 237–242.

Dysart, R. J. 1993. Impact of insect parasites and predators on grasshopper populations II. In: Cooperative Grasshopper Integrated Pest Management Project, 1993 annual report. Boise, ID: U.S. Department of Agriculture, Animal and Plant Health Inspection Service: 245–254.

Dysart, R. J. 1994. Impact of insect parasites and predators on grasshopper populations II. In: Cooperative Grasshopper Integrated Pest Management Project, 1994 annual report. Boise, ID: U.S. Department of Agriculture, Animal and Plant Health Inspection Service: 223–226.

Lakin, K. R. 1994. Assessment of the proposed release of Scelio parvicornis in the United States. Hyattsville, MD: U.S. Department of Agriculture, Animal and Plant Health Inspection Service, Plant Pro- tection and Quarantine, Biological Assessment and Taxonomic Sup- port, April 5, 1994. 34 p.

VII.9–4 VII.10 Ongoing Environmental Concerns

L. C. McEwen

Perhaps the greatest continuing environmental concern in many successful programs. The degree of grasshopper a Grasshopper Integrated Pest Management (GHIPM) reduction will probably be less than where liquid insecti- program is providing safeguards and protection for cide spray is applied, but the higher densities of grass- threatened and endangered (T and E) plant and animal hoppers remaining after the treatment often will be species. These problems complicate grasshopper control beneficial to the T and E species. programs and make them more costly but must be dealt with in a straightforward manner. Plenty of lead time Another possible option for protecting T and E species is should be allowed to identify species and habitats and to the timing of the grasshopper control program. This work out solutions with agencies responsible for T and E aspect can be explored for T and E insects and pollinators species’ protection and management. of T and E plants (also see chapter III.5). If the T and E insects are in the adult stage for a relatively brief period, Recognition of the fact that individual vertebrate animals then pest managers may conduct treatments safely before can vary greatly in their sensitivity to a given toxic or after the adult stage. chemical should help all workers understand that toxic exposure of the T and E species must be kept to a mini- For aquatic species, there are significant differences in mum. Toxic hazard is minor for mature animals lightly toxicity among the three chemicals. Acephate is much exposed to the current GHIPM pesticides—carbaryl, less toxic to fish than carbaryl or malathion (Johnson and malathion, and acephate—but is probably more of a fac- Finley 1980) and is referred to in other publications as tor for young animals (chicks, nestlings, amphibians, and practically nontoxic to fish. Acephate is highly effective larval fish). Any toxic mortality would be of concern against grasshoppers at the low application rate of 1.5 oz/ because species differ in their lower threshold of numbers acre (0.105 kg/ha) (U.S. Department of Agriculture of animals necessary for maintaining a viable population. 1987). Although acephate has been little used in coop- Those limits are not known precisely for each species, erative control programs, it could be an excellent alterna- but land managers should try hard not to cause unneces- tive to other pesticides where T and E fish are of concern. sary losses with toxic chemicals. Another safety factor for fish would be to use dry bait treatments because less chemical is used per unit area and In the larger picture, it would seem that concern for geo- there is much less potential for drift into aquatic habitat. graphic variants that have been given T and E status The entire problem of T and E species protection in should not be on the same level as for T and E species GHIPM programs could benefit from further research. that are the sole remaining population or individuals. Technically and legally, however, there is no distinction Indirect Effects on T and E Species at this time. The question of indirect effects of grasshopper control T and E species can be protected in several ways in a programs, primarily reduction or loss of the food base for rangeland grasshopper cooperative control program. birds, now comes up more frequently than potential toxic Nonspray buffer zones are one of the main tools (see effects. Colorado State University (CSU)-led studies chapter III.8). Width and size of buffer zones will vary have shown that when grasshopper availability is with the T and E species and on the outcome of consulta- reduced, birds generally switch to other insects or inver- tion with managing agencies. Carbaryl bait treatments or tebrates for food and maintain their nesting success and other dry baits, including biological control agents such populations (Miller 1993, Miller and McEwen 1995, as Nosema locustae and Beauveria bassiana, can be used Miller et al. 1994, George et al. 1995, Fair et al. 1995). safely much closer to the T and E species habitat or even Regarding the concern for peregrine prey effects, CSU with no buffer zone in some cases. investigators have shown that total bird population numbers do not decline following a grasshopper control pro- Baits and biologicals add expense and sometimes cause gram, even though some individual species might equipment problems when used but should be recognized decrease (George et al. 1995). Since peregrines prey on and accepted as important and necessary components of such a wide variety of avian species (DeWeese et al.

VII.10–1 1986, Hunter et al. 1988), the decline of one or two spe- spray as was done with Sevin® 4-Oil. One or more of cies should have no significant effect on their prey base. several species of concern are apt to be present in Use of dry baits, such as carbaryl bait, also could be a GHIPM treatment areas and should be treated as T and E safeguard since the baits are selective formulations and species if necessary in the opinion of the biologists and consequently leave many unaffected insects for avian land managers involved. food (Adams et al. 1994). Gallinaceous birds, such as prairie chickens and Nevertheless, each T and E species must be examined sharptailed grouse (Tympanuchus spp.), sage grouse individually for potential response to GHIPM treatments. (Centrocercus urophasianus), chukars (Alectoris chukar), The situation is such that T and E species and their habi- and wild turkeys (Meleagris gallopavo), also often are tats cannot be dealt with routinely by generalized proce- considered species of concern. The effects of grasshop- dures. Each T and E situation must be treated as a unique per control on the growth and survival of the young “case history,” although as knowledge is acquired, some chicks and poults is the primary question. More study is will be more standardized than others. needed on the effects of GHIPM programs on species of concern. New Chemicals and Biologicals Function of Wildlife in a GHIPM System New materials for range grasshopper control, such as Dimilin® (diflubenzuron) and Beauveria bassiana, will Scientists and land managers have made a lot of progress require close monitoring until their environmental safety in showing the role and benefits of wildlife, especially is determined. The two materials appear quite safe for birds, as important contributors to regulation of grasshop- terrestrial vertebrates, but final determinations cannot be per densities (Joern 1986, Fowler et al. 1991, Bock et al. made until the materials are applied in large-scale opera- 1992). However, the overall ecology of native wild ver- tional control programs. Aquatic effects are especially of tebrates in preventing insect pest outbreaks is virtually concern as well as Acridid (grasshopper) specificity and unexplored. The interrelationships of range condition, effects on nontarget invertebrates. Any other candidate vegetative cover types, native plants vs. introduced spe- chemicals and biologicals that are considered for GHIPM cies for reseeding (such as crested wheatgrass, Agropyron must also be closely examined for environmental effects cristatum), and associated wildlife populations need before being approved for large-scale use. much more investigation. Large expanses of crested wheatgrass become devoid of almost all the breeding Species of Concern avian species (Reynolds and Trost 1980). In the northern Great Plains, grasshopper outbreaks frequently originate State and Federal wildlife agencies in recent years have in crested wheatgrass, where grasshopper densities are endorsed a philosophy of giving attention to declining usually higher than on native grass range (Hirsch et al. species before they reach T and E status. If a declining 1988 unpubl., Kemp and Onsager 1994 unpubl.). This species can be managed for recovery before listing, man- fact should not be surprising because the lack of birds as agement efforts are simplified. Declining species may be grasshopper predators is coupled with >40 percent bare designated as “species of concern.” Some examples are ground (compared to <5 percent in native grassland the long-billed curlew (Numenius americanus), the west- (Dormaar et al. 1995), which is favored by many grass- ern burrowing owl (Athene cunicularia), and the ferrugi- hoppers for egg-laying. nous hawk (Buteo regalis). The curlews and burrowing owls use grasshoppers heavily, especially as a source of Range condition criteria are currently undergoing review protein and nutrients important for breeding and for feed- and revision (Task Group on Unity in Concepts and Ter- ing their young. The golden eagle (Aquila chrysaetos) is minology 1995). Land managers need to relate range another species of concern in some areas of the West and wildlife habitat use and populations to condition classes is a protected species. There is a need to conduct a study and to grasshopper population fluctuations. Improving of the response of nesting golden eagles to malathion range condition is a long, slow process, but range in good

VII.10–2 condition with a full complement of native wildlife can George, T. L.; McEwen, L. C.; Petersen, B. E. 1995. Effects of grass- reduce grasshopper population fluctuations in the central hopper control programs on bird populations in western rangelands. and northern Great Plains (McEwen 1987). Improving Journal of Range Management 48: 336–342. the condition of degenerated sagebrush (Artemisia spp.) Hunter, R. E.; Crawford, J. A.; Ambrose, R. E. 1988. Prey selection range found farther west is more difficult than improving by peregrine falcons during the nestling stage. Journal of Wildlife other range types, but it should be a long-term goal Management 52: 730–736. (McEwen and DeWeese 1987). New range management practices (Biondini and Manske 1996; Onsager, in press) Joern, A. 1986. Experimental study of avian predation on coexisting grasshopper populations (Orthoptera: acrididae) in a sandhills grass- should be examined for wildlife responses. land. Oikos 46: 243–249.

The status and function of wild vertebrates in relation to Johnson, W. W.; Finley, M. T. 1980. Handbook of acute toxicity of range condition also need more investigation. Basic chemicals to fish and aquatic invertebrates. Publ. 137. Washington, knowledge of range wildlife ecology connects with the DC: U.S. Department of the Interior, U.S. Fish and Wildlife Service. efforts to improve the vegetative cover on western range- 98 p. lands. Preventing the extinction of animal and plant spe- McEwen, L. C. 1987. Function of insectivorous birds in a shortgrass cies is the goal of conservation biology and will be a IPM system. In: Capinera, J. L., ed. Integrated pest management on benefit of better range condition. This will also be an im- rangeland: a shortgrass prairie perspective. Boulder, CO: Westview portant factor contributing to grasshopper management in Press: 324–333. an IPM system. McEwen, L. C.; DeWeese, L. R. 1987. Wildlife and pest control in the sagebrush ecosystem: ecology and management considerations. In: References Cited Onsager, J. A., ed. Integrated pest management on rangeland: State of the art in the sagebrush ecosystem. Bull. ARS–50. Washington, DC: Adams, J. S.; Knight, R. L.; McEwen, L. C.; George, T. L. 1994. Sur- U.S. Department of Agriculture, Agricultural Research Service: vival and growth of nestling vesper sparrows exposed to experimental 76–85. food reductions. The Condor 96: 739–748. Miller, C. K. 1993. Responses of nesting savannah sparrows to fluc- Biondini, M. E.; Manske, L. L. 1996. Grazing frequency and eco- tuations in grasshopper densities in interior Alaska. M.S. thesis. Ft. system processes in a northern mixed prairie. Ecological Applications Collins, CO: Colorado State University. 24 p. 6: 239–256. Miller, C. K; Knight, R. L.; McEwen, L. C. 1994. Responses of nest- Bock, C. E.; Bock, J. H.; Grant, M. C. 1992. Effects of bird predation ing savannah sparrows to fluctuations in grasshopper densities in inte- on grasshopper densities in an Arizona grassland. Ecology 73: rior Alaska. Auk 111: 960–967. 1706–1717. Miller, C. K.; McEwen, L. C. 1995. Diet of nesting savannah sparrows DeWeese, L. R.; McEwen, L. C.; Hensler, G. L.; Petersen, B. E. in interior Alaska. Journal of Field Ornithology 66: 152–158. 1986. Organochlorine contaminants in Passeriformes and other avian prey of the peregrine falcon in the Western United States. Environ- Onsager, J. A. [In press.] Suppression of grasshoppers in the Great mental Toxicology and Chemistry 5: 675–693. Plains through grazing management. Journal of Range Management. [Accepted for publication December 1, 1999.] Dormaar, J. H.; Naeth, M. A.; Williams, W. D.; Chanasyk, D. S. 1995. Effect of native prairie, crested wheatgrass (Agropyron cristatum) Reynolds, T. D.; Trost, C. H. 1980. The response of native vertebrate (L.)(Gaertn.) and Russian wildrye (Elymus junceus Fisch.) on soil populations to crested wheatgrass planting and grazing by sheep. Jour- chemical properties. Journal of Range Management 48: 258–263. nal of Range Management 31: 122–125.

Fair, J. M.; Kennedy, P. L.; McEwen, L. C. 1995. Effects of carbaryl Task Group on Unity in Concepts and Terminology. 1995. New grasshopper control on nesting killdeer in North Dakota. Environ- concepts for assessment of rangeland condition. Journal of Range mental Toxicology and Chemistry 14: 881–890. Management 48: 271–282.

Fowler, A. C.; Knight, R. L.; George, T. L.; McEwen, L. C. 1991. U.S. Department of Agriculture, Animal and Plant Health Inspection Effects of avian predation on grasshopper populations in North Service. 1987. Rangeland Grasshopper Cooperative Management Dakota grasslands. Ecology 72(5): 1775–1781. Program: final environmental impact statement. Washington, DC: U.S. Department of Agriculture, Animal and Plant Health Inspection Service. 221 p.

VII.10–3 References Cited–Unpublished

Hirsch, D. C.; Reuter, K. C.; Foster, R. N. 1988. Comparison of grasshopper population characteristics in relation to grassland types in western North Dakota in 1987. In: Cooperative Grasshopper Inte- grated Pest Management Project, 1988 annual report. Boise ID, U.S. Department of Agriculture, Animal and Plant Health Inspection Service: 244–252.

Kemp, W. P.; Onsager, J. A. 1994. Grasshopper population response to modification of vegetation by grazing. In: Cooperative Grasshopper Integrated Pest Management Project, 1994 annual report. Boise, ID: U.S. Department of Agriculture, Animal and Plant Health Inspection Service: 93–98.

VII.10–4 VII.11 Implications of Ecosystem Management and Information-Processing Technologies

W. P. Kemp, D. McNeal, and M. M. Cigliano

Ecosystem Management and Public Lands staggering in an effort to satisfy the need of policymakers to feel confident in presenting results for public viewing. A very large portion of the millions of rangeland acres in Add to this the challenge of a short interval between the 17 Western United States resides within the bound- problem identification and the time when action must be aries of what many refer to as the public land trust, or taken if it is to be effective for rangeland grasshopper federally managed lands. Voters have demanded that the IPM on public lands. It is clear that scientists and land public servants who manage these lands employ “ecosys- managers face an information-gathering and -processing tem management” to provide, among other things, a safe crisis. The remainder of this chapter will focus on ways food supply while not compromising natural resources that agencies can address this crisis that is already upon like clean air, clean water, productive soils, and the country. biodiversity. Private interests who lease grazing rights from the various public agencies charged with managing Present and Future IPM Technologies our national land treasure must comply with the public’s wishes regarding resource management issues or risk los- In spite of the information crisis faced with IPM on pub- ing the opportunity of using those public lands. lic lands, there are technologies available that agencies managing public lands can use in an attempt to comply At present, agencies involved in managing the natural with societal mandates. Other chapters in this Handbook resources on public lands are struggling to define just discuss global positioning system (GPS) and geographic what constitutes ecosystem management, how to manage information systems (GIS) for aircraft guidance (see sec- ecosystems whose limits do not agree with political or tion II) as well as for IPM in general (see chapter VI.9). ownership boundaries, and how to conduct such manage- The current role of modeling and decision support sys- ment with dwindling agency resources. For example, tems (DSS) also is discussed in the Decision Support there is general agreement throughout public land- Tools section. This chapter will focus on information management agencies that an ecosystem focus is desir- processing technologies and a new paradigm (example or able in managing the natural resources of public lands. model) in the context of IPM systems to be developed for There also is a nagging concern that agencies don’t have rangeland grasshoppers on public lands. a very clear vision of just how much information is necessary to meet national objectives. Furthermore, it is There are at least five areas of information-processing obvious that agencies will have to make natural resource technology that deserve additional attention in the devel- management decisions without complete information. opment of IPM systems for rangeland grasshoppers on Unfortunately, just what constitutes “enough” or “suffi- public lands, under the umbrella of ecosystem manage- cient” ecosystem management will likely emerge only ment. These are GPS, GIS, remotely sensed information, after and as a direct result of a series of court decisions. DSS, and networks. Three of the five areas—GPS, GIS, and remotely sensed information (see details in chapter Agencies cannot predict with absolute certainty what the VI.9) can be classified as technologies that assist land result of the ecosystem management consensus-building managers in collecting and storing information about the process will be, nor can they forecast the specific impacts ecosystems that they are responsible for managing. On ecosystem management will have on integrated pest man- the other hand, DSS and networks will be central to actu- agement (IPM) of public lands. The executive branch of ally processing the mountains of available information the Federal Government has provided some expected out- and developing the most appropriate management of a comes, at least in general terms (Gore 1993, National rangeland grasshopper problem on a particular piece of Research Council 1993). public rangeland.

In the case of rangeland grasshopper integrated pest man- Fortunately for public land-management agencies, there agement (IPM), many believe that the amount of infor- is a very competitive software and hardware market asso- mation needed to conduct management action (for ciated with GPS, GIS, and remote sensing technologies at example, chemical, biological, or cultural control) will be present. This competition is likely to continue well into

VII.11–1 the future. Such competition in the private sector of the GHIPM Project (U.S. Department of Agriculture, Animal U.S. economy will result in a steady and timely stream of and Plant Health Inspection Service [APHIS] 1987). products for use in collecting and storing information about the ecosystems that must be managed. Similar Instead of simply acknowledging that there are broad statements can also be made for the networking industry ecological differences in the Western United States, as everyone anticipates “information highways” of the agencies should use the concept of the ecoregion as a future. fundamental organizational paradigm. Bailey (1980) suggested that the regionalization (for example, fig. Perhaps the most serious challenge that agencies face in VII.11–1) that results from accepting this paradigm helps attempting to implement ecosystem management in gen- “(1) planning at the national level, where it is necessary eral, and rangeland grasshopper IPM in particular, is the to study management problems and potential solutions on development and maintenance of DSS. DSS such as a regional basis; (2) organization and retrieval of data Hopper, developed from funding provided by the Grass- gathered in a resource inventory; and (3) interpretation of hopper Integrated Pest Management (GHIPM) Project, inventory data, including differences in indicator plants must continually be updated and expanded to have any and animals among regions.” In our opinion, the capa- hope of processing the ecosystem information that is bilities that agencies have with GIS presently permit accumulating. In addition to defining who will be them to apply the ecoregion concept in ways that have responsible for the continued development of DSS, until now escaped scientists and land managers. agencies need coordinated planning to ensure that research emerging from Federal, private, and State labo- “Ecoregion” relates to the ability of the land to produce ratories will continue to support DSS improvements. goods and services that humans can use. Furthermore, historically sustainable activities related to grasslands We must note that, although technologies may be suffi- have to a large extent been molded by the prevailing con- ciently well developed for implementation and public ditions—expressed by ecoregion. For example, the dif- land-management agencies may be interested in adopting ferences in ranching styles and associated economics such technologies, costs will increase. This is true across the Western United States that economists have because of the significant increase in the information- been talking about are no doubt related to the fact that processing tasks presented by the implementation of eco- ranching has evolved in each region in response to the system management on public lands. The efficiencies of environmental limitations (again, expressed as operation with the equipment that is available today ecoregion). exceed even wild dreams of 10 years ago. Public land- management agencies are working feverishly to embrace Currently, Hopper (see VI.2) has been developed for only new technologies. There now is uncertainty whether the a part of the total area over which there is the opportunity resources will be forthcoming to do the job right. to use it. Furthermore, when land managers look at rangeland grasshopper economic injury levels (EIL) for Getting Organized widely separated areas, such as Wyoming and New Mexico, it is becoming more and more clear how impor- In this section, we offer some specific suggestions on tant the regional perspective can be. For example, recent how to coordinate future rangeland grasshopper IPM results suggest that it may take three to four times as with Federal land-management agencies. First, the con- many grasshoppers in New Mexico versus Wyoming cept of ecoregion—regional areas (fig. VII.11–1) with before management treatments would be justified eco- similar environmental resources, ecosystems, and sensi- nomically. In any case, whether agencies call them tivities to human impacts (Bailey 1980, Omernik 1987 ecoregions or rename them as management regions for and 1995) is useful for organizing information concern- the needs of APHIS, Plant Protection and Quarantine ing all aspects of grasshopper management. This is a (PPQ) activities, figure VII.11–1 represents a scale that is somewhat different use of the concept than was discussed a good first attempt to capture the variability across the in the environmental impact statement that governed the grasslands of the United States without overburdening people with too much detail.

VII.11–2 58

o n n e c t i c u t R r i v e

v i e R s r i o r T CENTRAL CORN BELT PLAINS CENTRAL CORN BELT PLAINS EASTERN CORN BELT S. MICHIGAN/N. INDIANA TILL PLAINS HURON/ERIE LAKE PLAIN HIGHLANDS NORTHEASTERN ZONE COASTAL NORTHEASTERN AND UPLANDS PLATEAU APPALACHIAN NORTHERN LAKE PLAIN ERIE/ONTARIO CENTRAL APPALACHIANS NORTH PLAIN COASTAL MIDDLE ATLANTIC PIEDMONT NORTHERN SOUTHEASTERN PLAINS BLUE RIDGE MOUNTAINS RIDGES AND VALLEYS CENTRAL APPALACHIAN SOUTHWESTERN APPALACHIANS CENTRAL APPALACHIANS WESTERN ALLEGHENY PLATEAU INTERIOR PLATEAU INTERIOR RIVER LOWLAND MISSISSIPPI ALLUVIAL PLAIN LOESS PLAINS MISSISSIPPI VALLEY PLAIN SOUTHERN COASTAL PLAIN SOUTHERN FLORIDA COASTAL 58 61 62 63 64 65 67 54 55 56 57 58 59 60 66 68 69 70 71 72 73 74 75 76 r

e 64 v

i 63

R s

e 76

m Albers equal area projection a 61 60 er J iv 62 R e e r t e n 65 iv a R 75 S h a n n a

69 v a S 70

67 66

n t

R i v e r

57 e

57 R

68 e

s v r

R v e i

n R

i 71 e

55 n e

h b

g i T o b e m

O T

v i I l l i n o R i s

R W

n o s

n i i o s v c

e 54

M s i s

e i

s s i p p v i i R 74 52 53

r 73 e iv R r i e

p v OZARK HIGHLANDS CENTRAL IRREGULAR PLAINS PLAINS GLACIATED MOUNTAIN NORTHERN PLAINS GLACIATED NORTHWESTERN PLAINS GREAT NORTHWESTERN NEBRSKA SANDHILLS PLAINS GREAT NORTHEASTERN PLAINS GLACIATED NORTHERN PLAINS WESTERN CORN BELT RED RIVER VALLEY WETLANDS MINNESOTA NORTHERN LAKES AND FORESTS NORTHERN CENTRAL HARDWOOD FORESTS NORTH DRIFTLESS AREA SOUTHERN WISCONSIN TILL PLAINS i 39 p i R 35 s i s r 42 43 44 45 46 47 48

i 39 40 41 52 38

49 51 53 u s 50

is o s 40

M s i

50 M 47 51 49 37 33 34 48 28 29

r 32

i 32

R 46 o d a

27 r o l

o C 45 30 31

(Revised 1993) r Rive s e 44 m r a 42 r e J e v

v i

r i R

e R

v d

i e s

R a

25 s

n a 25 a c i k l r

b 26

r A

e e

v i P

p R s e o

v e c

i 43

r e

21 v

e i n

i o

t 23 R s 16

i e

r w t

o a

u l l

o l

U.S. Environmental Protection Agency U.S. Environmental Protection e

s i

18 o i

n 23 G a r M 17 Corvallis Environmental Research Laboratory 41 20 19 22 16 Ecoregions of the Conterminous United States

r r COLORADO PLATEAUS SOUTHERN ROCKIES ARIZONA/NEW MEXICO PLATEAU ARIZONA/NEW MEXICO MOUNTAINS SOUTHERN DESERT WESTERN HIGH PLAINS SOUTHWESTERN TABLELANDS PLAINS CENTRAL GREAT FLINT HILLS CENTRAL OKLAHOMA/TEXAS PLAINS PLATEAU EDWARDS SOUTHERN TEXAS PLAINS TEXAS BLACKLAND PRAIRIES EAST CENTRAL TEXAS PLAINS PLAIN WESTERN GULF COASTAL SOUTH CENTRAL PLAINS MOUNTAINS OUACHITA ARKANSAS VALLEY BOSTON MOUNTAINS e

v e

i v

R i 27 24 28 31 20 21 22 23 25 26 30 32 33 34 35 36 37 38 29

n R

o o

m d

l a

a r r

e S o l v

C R

e k 10

n S 13 11 12 Ecoregions of the conterminous United States (adapted from a map in Omernik 1987 and 1995.) r 14 4 e

v i

R 6

a i

l 9

C 3

5 7

v i

R l o o r a o d C 1 8 6 1 PUGET LOWLAND WILLAMETTE VALLEY CASCADES SIERRA NEVADA SOUTHERN AND CENTRAL CALIFORNIA PLAINS AND HILLS CENTRAL CALIFORNIA VALLEY SOUTHERN CALIFORNIA MOUTAINS EASTERN CASCADES SLOPES AND FOOTHILLS COLUMBIA PLATEAU BLUE MOUNTAINS SNAKE RIVER BASIN/HIGH DESERT BASIN AND RANGE NORTHERN SOUTHERN BASIN AND RANGE ROCKIES NORTHERN AND FOOTHILL PRAIRIES VALLEY MONTANA MIDDLE ROCKIES WYOMING BASIN MOUNTAINS AND UINTA WASATCH 7 4 5 1 COAST RANGE 2 3 8 9 6 11 15 17 18 10 13 14 12 16 19 Figure VII.11–1—

VII.11–3 The ecoregion concept is useful for exchanging information about environmental conditions, plant production, Organization scheme for a Final Environmental ranching, and grasshopper ecology and management Impact Statement for a Rangeland Grasshopper (from hatching to outbreak frequency and probability and Cooperative Management Program more). There is a credible argument for the use of the concept of ecoregion as a framework for the development Level 1: Ecoregions—regional variations in cli- of future rangeland grasshopper cooperative management mate, vegetation, and landform. This is program final environmental impact statements (FEIS’s). the basis for organizing what agencies The ecoregion concept also has potential application for know as well as what and how agencies other pest-related issues (for example, noxious weeds) will manage. with which APHIS, PPQ and Federal land-management agencies must deal. Level 2: Things that are likely to be different by ecoregion and that should be considered In the development of any future FEIS activities, pest in any future activities related to the managers first should organize rangeland grasshopper Rangeland Grasshopper Cooperative IPM activities to be responsive to the situations recog- Management Program FEIS (this list is nized within each ecoregion. Next, agencies should not meant to be all-inclusive): acknowledge that IPM is the collection of options (including no action) and philosophies most appropriate • Grasshopper community species for addressing grasshopper management. Considering composition, the variation in grassland vegetation and climate depicted • Likelihood of grasshopper outbreaks, in figure VII.11–1 and associated variations in grasshop- • Spatial extent of grasshopper per populations (for example, Kemp et al. 1990), it is outbreaks, very unlikely that all management options will be equally • General insect–animal community viable (as viewed by environmentalists, economists, and composition, the public) or of constant efficacy across the rangelands • Grassland plant community of the 17 Western United States. If this approach to man- composition, agement is acceptable, then there is a logical manner for • Forage production on grasslands, studying and determining what to emphasize in terms of • Economics of ranching and farming IPM components at the ecoregion level. (and thus land use and human population density), Using this approach as an example, the tabulation in the • Economics of grasshopper control and right column illustrates one way to organize an FEIS. EIL, • Endangered species, • Soils (and thus water and pesticide movement), and • Water resources.

VII.11–4 1993 Grasshopper Survey

Density/yd2 0 Ð 3 0+ Ð 8 8+ Ð 15 15+

Figure VII.11–2—Locations in the 17 Western United States where (starting in 1993) rangeland grasshoppers were sampled annually for density and species composition by USDA, APHIS, PPQ and cooperators for the Grasshopper Common Dataset Project. Colors indicate grasshopper density at each location in 1993.

VII.11–5 The ecoregion paradigm, in addition to being politically 4. Given 1–3, the regionality provided by the ecoregion and environmentally acceptable (see Gore 1993, National concept has great potential for clarifying the goals and Research Council 1993), can provide Federal land man- objectives of research that Federal land-management agement agencies and APHIS, PPQ with a powerful tool agencies and APHIS, PPQ should obtain through con- for organizing and interpreting research results relative to tracts and cooperative ventures. rangeland grasshoppers. For example, discussions among a number of GHIPM Project participants and References Cited APHIS, PPQ staff eventually resulted in the initiation of the Grasshopper Common Dataset (GCD) during 1993. Bailey, R. G. 1980. Description of the ecoregions of the United States. Scientists now are monitoring rangeland grasshopper Misc. Publ. 1391. Washington, DC: U.S. Department of Agriculture, Forest Service. 77 p. communities annually at more than 1,500 locations throughout the 17 Western United States (fig. VII.11–2). Gore, A. 1993. Earth in the balance: ecology and the human spirit. Results from ongoing research by GHIPM Project coop- New York: Plume. 408 p. erators, with data from the GCD, will tell to what extent grasshopper communities are sensitive to the ecoregion Kemp, W. P.; Harvey, S. J.; O’Neill, K. M. 1990. Habitat and insect boundaries shown in figure VII.11–1. Given that scien- biology revisited: the search for patterns. American Entomologist 36: 44-48. tists are able to identify ecological boundaries that are in some way meaningful to the insects, scientists and land National Research Council. 1993. A biological survey for the Nation. managers should apply this concept to assist them in Washington, DC: National Research Council, National Academy organizing the way that they think about things like Press. 205 p. rangeland grasshopper management on grasslands west Omernik, J. M. 1987. Ecoregions of the coterminous United States. of the 100th meridian of the United States. Annual of the Association of American Geographers 77: 118–125.

In summary, the four main points that we wish to empha- Omernik, J. M. 1995. Ecoregions: a spatial framework for environ- size are mental management. In: Simon, T.; Davis, W., eds. Biological assessment and criteria: tools for water resource planning and decision 1. GPS, GIS, remote sensing, networking, and DSS will making. Chelsea, MI: Lewis Publishers: 49–62. be necessary for ecosystem management of public U.S. Department of Agriculture, Animal and Plant Health Inspection rangelands. Service. 1987. Rangeland Grasshopper Cooperative Management Program: final environmental impact statement. Washington, DC: 2. The ecoregion concept is useful, deserves additional U.S. Department of Agriculture, Animal and Plant Health Inspection consideration by Federal land-management agencies Service. 221 p. and APHIS, PPQ, and could serve as a useful paradigm for organizing future environmental impact statements related to rangeland grasshoppers (and possibly other insects).

3. By accepting the ecoregion concept, agreeing that IPM is the basis for all grasshopper management, and accepting that IPM consists of all possible alternatives and philosophies as above, agencies eventually could develop ecoregion-specific IPM prescriptions for rangeland grasshopper management.

VII.11–6 VII.12 Rangeland Environmental Amenities and Grasshopper Management Programs

Melvin D. Skold and Andrew W. Kitts

Range Ecosystems The biodiversity of rangelands contributes to the intangible products mentioned in the National Research Coun- Rangelands are increasingly recognized as important for cil (1994) report. Recognition of the importance of their environmental and recreational amenities. Because biodiversity arises for several reasons: (1) morality, they are managed much less intensively than many other (2) esthetics, (3) economics, and (4) “biological types of agricultural lands, rangelands are seen to repre- services.” sent closer approximations to natural ecosystems. Rangelands are managed for a variety of outputs; in Increasingly land managers are learning of the effects of recent years, the contribution of natural rangeland sys- the impacts of management or lack of management on tems to biological diversity has become increasingly the ability for various species to survive. Some assert recognized. that mankind has a moral obligation to protect fellow creatures. Social awareness has also made managers and Rangelands provide two major values, those associated others aware of the need to protect spaces, natural sys- with use (use values) and those realized in the absence of tems, and historic sites. In addition to the value of direct use (existence and option or nonuse values). The present consumptive and nonconsumptive uses, range- major commercial use (use values) of rangelands is live- lands also possess esthetic values, and other economic stock grazing to produce food, fiber, and draft animals. potentials exist. Potentially these natural systems include Other, less significant, commercial uses such as wild yet-to-be-identified goods that could be of value to game and bird hunting also are associated with rangeland people. Finally, ecosystems are important components of habitats. In addition, rangelands are viewed as important natural cycles affecting the gaseous composition of the contributors to watersheds: because rangelands usually atmosphere; genesis, fertility, and stability of soils; dis- have lower rates of soil erosion than cropland, they posal of wastes; cycling of nutrients; and natural control enhance water quality. Further, the natural system that of pathogenic and parasitic organisms (West 1993). exists on well-managed rangelands makes them increasingly recognized as places for nonconsumptive wildlife A healthy range is recognized as one in which the integ- associated recreation. rity of the soil and ecological processes of the rangeland ecosystem are sustained (National Research Council Rangelands also produce intangible products (or nonuse 1994). Whenever management intervenes in the natural values) that are the result of use. These products include processes, for whatever reason, the impact of those inter- natural beauty, open space, and the mere existence as a ventions on the rangeland’s ability to sustain commercial natural ecosystem (National Research Council 1994). as well as intangible products must be considered. Others emphasize biological diversity and the associated Rangeland grasshoppers also can disrupt the natural eco- potential array of products and services as a distinct system in two ways. First, grasshopper infestations can intangible product (West 1993). In contrast to use val- reach plague proportions. Serious and widespread out- ues, nonuse values occur almost entirely outside the mar- breaks can lead to soil erosion and reductions in water ket system. However, methods are evolving to quantify quality and make it difficult—if not impossible—for the and assign monetary value to these existence values. As range to recover to its original state. Major infestations with use values, the costs and/or trade-offs associated of grasshoppers destroy cover for ground-nesting birds with nonuse values can be compared to the estimated and mammals and damage the habitat for other wildlife. benefits (Bishop and Welsh 1992.) The desire to protect the range ecosystem and adjacent croplands was an important part of the rationale for initi- Rangelands possess attributes that give them potential for ating the publicly assisted rangeland grasshopper control biodiversity. Since they have not been “put to the plow,” programs that exist today. rangelands are attributed value as a natural system. Fur- ther, rangelands cover vast areas, often contiguously, and Second, grasshoppers are recognized as an integral and thereby possess the scale necessary for biodiversity of necessary part of a range ecosystem. Grasshoppers and communities, ecosystems, and landscapes (West 1993). other rangeland insects are an important part of the food

VII.12–1 chain of some birds and mammals. Some species of found 80 percent of those spending summer vacations in grasshoppers are beneficial, feeding on plant forms that the Steamboat Springs area indicated that ranch open are not consumed by other users of the range. Because space added significantly to their willingness to pay for grasshoppers cut off vegetation as well as consume it, summer visits. Willingness to pay for ranch open space they create litter that becomes an important part of the averaged about $20 per day (Walsh et al. 1993). nutrient cycle on rangelands. The strategy for managing rangeland grasshoppers has to be one of maintaining bal- Many of the biological–physical–management interac- ance within range ecosystems. tions associated with rangeland biodiversity are yet to be understood (West 1993). Consequently, very little has The Grasshopper Integrated Pest Management (GHIPM) been done to evaluate the contributions of rangelands to Project recognized the potential environmental costs asso- biodiversity. Yet, under the Forest Management Act of ciated with applying grasshopper management programs. 1976 and the Surface Mining Control and Reclamation One component addressed the safe use of grasshopper Act of 1977, rangelands must be managed for biodi- management programs around threatened and endangered versity. Intangible values are reflected in policy plant species (Tepedino and Griswold 1993 unpubl.). directives even if quantification of those values has not Another chapter (III.6) in the environmental monitoring occurred. and evaluation section of the User Handbook evaluates the effects of grasshopper treatments on wildlife and Reported here is an example of how rangeland environ- aquatic species. The economics component of the mental amenities can be evaluated. Chapter VI.3 of this Project developed procedures to make estimates of the Handbook discusses the method of estimating the eco- environmental costs of control programs. This valuation nomic loss to ranchers from an uncontrolled grasshopper recognizes, as the reader shall subsequently see, that fish outbreak. Applying chemical treatments reduces dam- and wildlife possess a value for recreation that considers ages for the livestock grazer, and the damage reductions both nonconsumptive (bird watching, photography, are the benefits of grasshopper controls. Pest managers hiking) and consumptive (fishing, hunting) forms of also can estimate the economic loss if grasshopper con- wildlife-associated recreation. trol activities deplete wildlife populations. Figure VII.12–1 shows the flow of events. Grasshopper program managers have been conscious of possible environmental side effects, undesired and ben- If grasshopper management programs deplete wildlife eficial, from these programs. Chemical applications may populations, a reduction in the wildlife base will result affect populations of some nontarget insect species as in fewer people participating in wildlife-associated well as grasshoppers. Treatment program managers warn recreation. Because people place an economic value keepers of commercial insects so that those populations on recreation, less recreation means an economic loss. are protected. Managers of treatment programs take care Investigators link the economic evaluation of wildlife to spray chemicals under conditions that minimize drift depletion to grasshopper management and take the eco- and to refrain from applying certain chemicals near nomic losses from wildlife-associated recreation as a water. measure of the portion of the environmental costs of the grasshopper treatment programs. Evaluating Losses in Wildlife-Associated Recreation Calculations can start with the net economic values of wildlife-associated recreation estimated by Hay using Economists have made estimates of the value of some of willingness-to-pay techniques (1988a and b). Using the the nontraditional outputs from rangelands (Bernardo et net economic value estimates for specific regions, it is al. 1992, Kitts 1992, Loomis et al. 1989, Standiford and possible to make estimates of the reduction in consump- Howitt 1993, Young et al. 1987). Most of these studies tive and nonconsumptive forms of wildlife-associated have focused on consumptive and nonconsumptive forms recreation resulting from a decrease in the wildlife of wildlife-associated recreation. However, a recent resource base. Colorado study estimated the value of open space. It

VII.12–2 life Service conducts periodic surveys of fishing, hunting, Grasshopper management and wildlife-associated recreation. The year for which ➔➔➔ the most recent survey data are available is 1985. Many factors determine the likelihood that an individual will Wildlife depletion participate in wildlife-associated recreation. For discussion in this chapter, we are primarily interested in one variable—the effects of the wildlife resource base on the Reduction in wildlife-associated recreation probability of participation. If the wildlife resource base declines, we expect that the rate of participation in wildlife-associated recreation also will decline. Since grass- Loss of net economic value hoppers and grasshopper treatments affect the habitat of wildlife, a measure of the wildlife resource base is Figure VII.12–1—Sequence of events from grasshopper management habitat. to loss of value. For hunting and nonconsumptive forms of wildlife- associated recreation, the amount of participation was Analyzing the information reveals how participation in sensitive to changes in the wildlife resource base. Fish- wildlife-associated recreation depends on demographic ing was not responsive to an estimate of changes in the variables, price (cost of participating in recreation) and fishing resource base. For hunting, a reduction of the wildlife resource base. Managers can use analyses 1 percent in the range habitat of wildlife (for example a for each type of wildlife-associated recreation (fishing, 1-percent reduction in the capacity of a range to support hunting, and nonconsumptive recreation) in the States for game wildlife) results in a 3.2-percent reduction in hunt- which control of rangeland grasshoppers is a problem. ing participation. Similarly, a 1-percent reduction in the rangeland wildlife base results in a 2.9-percent reduction The economic analysis involves the last two linkages of in participation in nonconsumptive forms of wildlife rec- figure VII.12–1. Potential wildlife depletion results in a reation. reduction in wildlife-associated recreation that, in turn, results in a net economic loss. This loss is a measure of a Table VII.12–1—Net economic values per day of part of the potential environmental costs associated with wildlife-associated recreation, by recreational grasshopper management programs. activity in the eight-State region1

Potential Environmental Costs Net economic value Activity (dollars/day) Table VII.12–1 shows Hay’s net economic values for wildlife-associated recreation. These are the average net Hunting economic values for the eight States included in and sur- Deer $35 rounding the GHIPM demonstration sites. The net eco- Elk $36 nomic values are from surveys designed to determine Waterfowl $20 how much participants value a day of recreation in these activities. Fishing $11

The next step to estimating the potential loss in wildlife- Nonconsumptive $22 associated recreation resulting from grasshopper management programs is to look at the relationship between the 1Idaho, Montana, Nevada, North Dakota, Oregon, wildlife resource base and the amount of participation in South Dakota, Utah, Wyoming. Source: Hay (1988 a and b). wildlife-associated recreation. The U.S. Fish and Wild-

VII.12–3 The statistical equations give estimates of the number of Potential Recreation Losses participants in each wildlife-associated recreation activity. In this chapter, we focus on how wildlife-associated The economic losses associated with changes in the wild- recreation changes in response to changes in the resource life resource base are only potential losses. The environ- base. Table VII.12–2 shows the base level estimate of mental monitoring component of the GHIPM Project has the number of hunters in the eight-State region, their not found adverse effects on wildlife resulting from use expenditures, participation days, and the net economic of grasshopper control programs. Approved treatment value from hunting in the region. options are the result of careful evaluation and selection to determine materials and methods which minimize the The table also shows the potential impact of a 1-percent threat to the environment. When there are grasshopper decline in the game wildlife resource base and the associ- treatments, these precautions to minimize the environ- ated economic impact. We can interpret the analysis two mental damage apparently are successful. So long as the ways. A 1-percent increase in the wildlife resource base first linkage in figure VII.12–1 remains zero, meaning would result in an increase of the same magnitude in par- grasshopper treatments do not result in wildlife depletion, ticipation, expenditures, hunting days, and net economic the economic losses from reductions in wildlife- value, as would a 1-percent decrease. Thus, if the use of associated recreation are also zero. However, should a grasshopper treatment program reduces the wildlife damages to the wildlife resource base occur, the changes resource base, we can measure the cost (loss in net eco- in net economic value due to wildlife-associated recre- nomic value). Conversely, if grasshoppers destroy the ation can be estimated by applying this procedure. habitat for wildlife and a reduction in game wildlife occurs, we also can estimate the potential losses from less Conclusions hunting on grasshopper-damaged rangeland. With increased understanding of the linkages and rela- Using the estimated equations for nonconsumptive forms tionships present in rangeland ecosystems, it will be pos- of wildlife recreation, table VII.12–3 shows the base eco- sible to quantify more of the identified benefits from nomic activity and potential losses if a grasshopper inva- rangeland biodiversity and other intangible values. Until sion reduces the wildlife resource base. As with hunting, that time, rangeland management and actions taken to nonconsumptive wildlife-associated recreation also may control rangeland pests must proceed with the best avail- suffer if an uncontrolled grasshopper outbreak reduces able understanding of the results from those management the wildlife resource base. interventions.

Table VII.12–2—Hunting: Effect of reduced wildlife resources on the number of participants and trip-related expenditures and on participation-days and net economic value

Wildlife resource Number of Trip-related Participation- Net economic level participants expenditures days value

Thousands $ million Thousands $ million

Base level 790,000 $191.2 11,847 $355.4

1% decline –25 –6.1 –371 –11.1

VII.12–4 Table VII.12–3—Nonconsumptive: Effect of reduced wildlife resources on number of participants and trip- related expenditures and on participation-days and net economic value

Wildlife resource Number of Trip-related Participation- Net economic level participants expenditures days value

Thousands $ million Thousands $ million

Base level 1,501 $253.7 15,009 $330.2

1% decline –43 –7.3 –429 –9.4

National Research Council. 1994. Rangeland health: new methods to References Cited classify, inventory and monitor rangelands. Washington, DC: Com- mittee on Rangeland Classification, Board of Agriculture. Bernardo, D. J.; Engle, D. M.; Lochmiller, R. L.; McCollum, F. T. Standiford, Richard B.; Howitt, Richard E. 1993. Multiple use man- 1992. Optimal vegetation management under multiple use objectives agement of California’s hardwood rangelands. Journal of Range Man- in Goss timbers. Journal of Range Management 45: 452–469. agement 46: 176–182. Bishop, Richard C.; Welsh, Michael P. 1992. Existence values in Walsh, R. G.; McKean, J. R.; Mucklow, C. J. 1993. Recreation value benefit–cost analysis. Land Economics 68: 405–417. of ranch open space. Report to the Routt County (CO) Board of Com- missioners. Ft. Collins, CO: Department of Agricultural and Resource Hay, M. J. 1988a. Analysis of the 1985 National Survey of Fishing, Economics, Colorado State University. 43 p. Hunting, and Wildlife-Associated Recreation—net economic recreation values for deer, elk and waterfowl hunting and bass fishing. Rep. West, Neil E. 1993. Biodiversity of Rangelands. Journal of Range 85-1. Washington, DC: U.S. Department of the Interior, U.S. Fish and Management 46: 2–13. Wildlife Service. 23 p. Young, John S.; Donnelly, Dennis M.; Sorg, Cindy T.; Loomis, John Hay, M. J. 1988b. Analysis of the 1985 National Survey of Fishing, B.; Nelson, Louis J. 1987. Net economic value of upland game hunt- Hunting, and Wildlife-Associated Recreation—net economic values ing in Idaho. Resour. Bull. RM-15. Ft. Collins, CO: U.S. Department of nonconsumptive wildlife-related recreation. Rep. 85-2. Washing- of Agriculture, Forest Service, Rocky Mountain Forest and Range ton, DC: U.S. Department of the Interior, U.S. Fish and Wildlife Experiment Station. 23 p. Service. 16 p.

Kitts, Andrew W. 1992. Economic value of some external costs for References Cited–Unpublished grasshopper control. M.S. thesis. Ft. Collins, CO: Colorado State Uni- versity, Department of Agricultural and Resource Economics. 91 p. Tepedino, Vincent J.; Griswold, Terry L. 1993. Pollination biology of threatened and endangered plants. In: Cooperative Grasshopper Loomis, John; Donnelly, Dennis; Sorg-Swanson, Cindy. 1989. Com- Integrated Pest Management Project, 1993 annual report. Boise, ID: paring the economic value of forage on public lands for wildlife and U.S. Department of Agriculture, Animal and Plant Health Inspection livestock. Journal of Range Management 42: 134–138. Service: 181–189.

VII.12–5

VII.13 Grasshopper Communities and Methodology

Anthony Joern

Grasshopper populations do not exist in an ecological cost–benefit ratios ultimately determine the value of vacuum. Instead, individual species populations interact studying community relationships. with several other species, other individuals, other herbivores, a range of potential host plants and many natural As I list accepted methods to evaluate grasshopper com- enemies. In western North America, 30 to 50 grasshop- munities, I will stress the difference between merely per species may coexist, and each may respond individu- describing community composition (species identities) ally to environmental change. Although science’s and understanding mechanisms driving species interac- interest lies mainly in the ecology and population dynam- tions and coexistence. IPM measures interrupt dynamic, ics of a single or a few species, one species cannot often subtle, ecological interactions within and among exempt itself from a network of interactions among all species. Until we work out the impact of these key inter- species that are present. Consequently, the grasshopper actions for many species combinations in detail, species community becomes a central focus in any rational inte- lists alone provide little insight into future system dynam- grated pest management (IPM) project. ics surrounding IPM efforts.

Communities are significantly more complex to evaluate Community Descriptions: and study than single-species populations. Manipulating List of Grasshopper Species Present one small component of the community network (e.g., of one or a few species) may not evoke the desired, long- A list of grasshopper species is the simplest description term control objectives. Consideration of only one or a of a community and is required in any community-level few species may lead to unnecessarily short-term solu- assessment. A good description includes the relative tions or even to unexpected problems. Besides problems abundance and absolute density of individual species in a associated with community complexity, species assem- community. Density is important because the number of blages vary greatly from year to year at the same site and individuals that are available to interact determines, at vary even more dramatically among sites. Scientists least in part, what really happens. require descriptive and analytical methodologies to clearly devise and assess community management prac- Based on past studies, experts can sometimes develop tices. Scientists also must simplify the scope of the insights regarding community dynamics from such lists— problem without sacrificing important connections that if certain conditions and species are present. Shifts in prescribe creative solutions. species composition among years or among sites suggest that different grasshopper species react differently to In this section, I summarize simple, standard approaches changing environments. Such variation in the response and methodologies for describing communities and for to different environmental conditions indicates that either assessing the importance of key interactions. Some of the community shifts from one state to another or that the these methods are best for sporadic evaluation of random internal dynamic interactions among species shift. Con- sites on a hit-or-miss basis. Others are designed for sequently, the same IPM management practice employed developing long-term understanding at sites that are regu- under different conditions may produce different long- larly monitored for potential grasshopper problems. term responses depending on the state of the community. Government agencies and private organizations that manage the same large tract over many years can expect to Sampling efficiency can vary with habitat type and its develop comprehensive, community-based IPM pro- three-dimensional structure as well as overall grasshop- grams. But individual ranchers with only intermittent per densities. Typical methods include sweeping some grasshopper problems and few resources cannot. As a predetermined number of times or counting grasshoppers result, managers must select which of the following at stationary sample sites (e.g., the “ring technique” of approaches to community evaluation meets their situa- Onsager and Henry 1977, Thompson 1987). Berry et al. tion. Complete annual censuses and evaluations of review appropriate sampling methods and their justifica- environmental conditions are the cornerstones of commu- tion in chapter VI.10 of this handbook. Remember, nity studies. These require significant effort, and that in obtaining lists of species’ relative abundances, the

VII.13–1 accurate sampling of rare species is the biggest problem. range of readily consumed plant species will be similar More samples will reduce the chance of missing rare spe- among sites. Based on use of both food plants (Joern cies. To estimate a sampling intensity that will detect 1979a, 1983) and microhabitat resources (Joern 1982), most of these species at your site, plot the cumulative community level patterns emerge that may help a number of grasshopper species collected against some manager make decisions (Joern 1979a,b, 1986a). The measure of sample intensity (number of individuals col- usefulness of such an approach for developing sound lected, number of sweeps, number of rings examined, grasshopper IPM tactics is idiosyncratic and case-specific number of transects, area sampled, and number of habitat at this time. types sampled). Figure VII.13–1 illustrates a reasonable sampling schedule. In designing sampling plans, be Using Statistics To Estimate Species aware that you will probably encounter some unrecorded Replacements and Community species if new habitat types are included. Because of Associations this, plan to sample all habitat types found in the area in the proportion that they occur in the environment. Species replacements and community associations along environmental gradients can be identified using standard What rules-of-thumb emerge from species lists? Many multivariate statistical techniques (e.g., discriminant species thrive only in areas with open bare areas (e.g., function analysis, principle components analysis, Ageneotettix deorum). Other species (e.g., Paropomala detrended correspondence analysis) or some combination wyomingensis) require significant vertical structure such of the statistical techniques developed for ordinating as that provided by bunchgrasses. Still other species communities (Gauch 1982). As a technique, ordination (e.g., Melanoplus sanguinipes) occupy a variety of mi- simplifies multiple species associations by representing crohabitats, so that little insight can be gained just by the relationships in fewer dimensions using mutivariate knowing what microhabitats exist at a site. Similarly, descriptive statistics. By using these techniques, you can even among grasshopper species that eat many plants, the identify the combinations of species that tend to occur together (and their relative abundances) in association with key attributes of the environment such as vegetation Cumulative no. of species type or soil moisture (fig. VII.13–2). Such community analyses allow you to simplify the community associations along a spatially varying environmental gradient. Be aware of the correlational nature of these results from these analyses. The patterns that you uncover will fully depend on what you include in your initial sampling design. If you add species or sites with different combinations, the ultimate patterns may shift. Ordination provides a refined fit between grasshopper community composition and some environmental gradient, but you cannot identify dynamic and causal relationships between the two features by using this approach.

Plotting Against an Environmental Gradient.—You can readily visualize species replacements along gradi- Sampling intensity * ents by plotting the change in the abundance (or relative abundance) of each species along some environmental Figure VII.13–1—The number of species sampled is dependent on the sampling intensity. To obtain a good estimate of the number of gradient (fig VII.13-2a). In this hypothetical analysis, I species at a site, sampling intensity should equal that indicated with an assess a series of independent sample sites as in number 1 asterisk, near the asymptote for the entire assemblage. If sampling in- above (a list of grasshopper species). Then, on a species- tensity is less than this point, many rare species will likely be missed. by-species basis, I plot the abundances (or relative abun-

VII.13–2 dances) along the gradient. By comparing these plots puter algorithms help put boundaries around species among species, you can identify possible environmental combinations from each location, largely based on conditions at your site best suited and worst suited for changes in relative abundances rather than in response to each species. In addition, you can compare responses of massive replacement of individual species. Remember, multiple species along the same gradient. these boundaries of species composition represent “probability boundaries” and much overlap typically exists in Multivariate Ordination Techniques.—Species asso- grasshopper species composition among adjoining com- ciations can be identified using standard, multivariate munities or even when comparing sites some distance ordination techniques (fig. VII.13-2b). While these tech- away. As a warning: many users of this technology tend niques typically require commercially prepared computer to become typological in describing communities and software, the analyses are readily accessible, even on often confuse pattern with a dynamic process. For laptop computers. Standard references exist to help the example, I foresee some managers ordinating grasshop- user understand both the statistical guts of the analysis as pers from a group of sites and then prescribing specific well as providing insights to interpreting results (Cornell management options for those assemblages in group A Ecology Programs discussed in Gauch 1982). The com- versus group B or C and so on. The assumption that all

A B

Species abundance Axis 2

4 A 1 6 10 2 8 5 7 9 B 3

ABCDEFGH Axis 1 Environmental gradient

Figure VII.13–2—A: Hypothetical distribution of species along some environmental gradient based on sampling at 8 sites (A–H) along a transect. Each curve indicates the distribution along this gradient for a hypothetical grasshopper or plant species. For example, species 4 does best at site C but does not exist at site E while species 3 does not do particularly well at any site but is found along the entire gradient. B: This multivariate distribution can be “boiled down” into a simpler relationship using ordination techniques following those outlined in Gauch (1982). Each of these new axes (1 and 2) represent a composite of multivariate data. The points indicated in B represent the average position for each species indicated in A for the two multivariate resource axes developed from a composite of environmental variables. The groupings of species indicated by the dashed lines suggest species that react to environmental conditions in the same fashion. Examples of gradient analyses of grasshopper species along a topographic gradient in Montana are presented in Kemp et al. (1990) and Kemp and O’Neill (1990).

VII.13–3 sites exhibiting type A species associations also categori- Consequently, an efficient experimental approach cally exhibit the same underlying dynamics is unfounded. requires a strong conceptual framework so that science can simultaneously evaluate key competing possibilities Unless a conceptual framework exists that predicts and that investigators can reject alternatives based on unique, species-specific relationships, the results will not experimental results. The conceptual framework identi- explain why specific patterns emerge. For example, fies alternate hypotheses. By simultaneously testing grasshopper species assemblages often change predict- competing explanations of community pattern and pro- ably as the species composition of the plant community cess through experimentation, the manager can rapidly changes (see chapter IV.3). What dynamic relationship narrow the options. Then it becomes possible to uncover exists between the two components of this analysis to the best explanations upon which to base management explain the results? Unfortunately, insufficient informa- options. Despite the difficulties and cost, I strongly tion exists to tease apart such relationships, even if the believe that the intense effort required to uncover site- pattern is very strong. Sometimes specific theories exist specific dynamics using controlled manipulations will that predict particular species responses in abundance or pay off, in the long term, for grasshopper IPM managers. in association with specific habitats. In these situations, Examples of sites that should profit from intensive stud- additional insights regarding dynamic, causal mecha- ies include public lands and large private holdings with nisms might emerge from pattern analysis, but this notion constant or predictable land-use practices and a history of still requires experimental testing to uncover the underly- grasshopper problems. If managers feel insecure about ing reasons for the relationships fully. Scientists must performing all of the above work by themselves, they base management options on processes driving commu- should allocate some management funds to contract for nity dynamics, not on easily measured patterns. This fact research by competent scientists. is unfortunate because scientists can more readily establish measures of pattern than uncover the underlying A current example illustrates the above process. A con- dynamic mechanisms. ceptual framework that defines alternate views of the problem, combined with experimental manipulation and Using Controlled Manipulations To coupled with appropriate comparisons and descriptive Uncover Site-Specific Dynamics analyses, allows recognition and interpretation of the dynamic interactions that regulate community-level pro- Experimental manipulation of species interactions can cesses. As a general framework, the alternatives include provide powerful community level insights into the “top-down” versus “bottom-up” processes (Hunter et al. dynamic forces that organize communities. However, 1992). As herbivores, grasshoppers occupy an intermedi- the effort is great. From an IPM framework, subtle shifts ate trophic (nutrition) position in the food web, with food in species composition that changes in the underlying in- plants below them and natural enemies (e.g., parasitoids, teraction dynamics may provide the key for developing invertebrate and vertebrate predators, or fungal, bacterial, the correct management strategy. After all, those IPM or viral pathogens) positioned above them. practices that work in concert with naturally occurring dynamic processes will most likely lead to long-term suc- What major forces limit grasshopper populations in this cess. However, uncovering the specific nature and food web? From a control standpoint, this information strength of interactions among species, including their provides the clue to appropriate management planning. impact on resulting population densities and community Bottom-up forces can arise from insufficient nutrients structure, will require experimental manipulations under either when grasshoppers compete for limited food or field conditions. Standard experiments that might when time constraints interfere with feeding and diges- uncover these relationships are time consuming and tive capability. Top-down forces can arise from the complex. actions of natural enemies. Other chapters of the Grass- hopper Integrated Pest Management User Handbook provide detailed examples of each type of interaction.

VII.13–4 Descriptive studies cannot untangle this set of potential may lead to unexpected consequences from control interactions, but manipulative experiments can. In fact, efforts if we ignore rare but otherwise functionally under natural conditions, bottom-up (Belovsky and Slade important taxa. Both species lists and more complicated 1995) and top-down (Joern 1986b, 1992 ) forces operate statistical descriptive techniques of grasshopper commu- simultaneously, and either one can drive the interactions nities will provide some guidelines, but neither will pro- and can thus determine the final densities of coexisting vide direct insights about dynamic relationships. grasshoppers (Belovsky and Joern 1995). More impor- Because effective control will result in permanent or at tantly, reciprocal indirect effects of species on each other least long-lasting alteration of species interactions, scien- can potentially be more important than the direct interac- tists would like to understand the dynamics of these inter- tions. Scientists can see such responses only through actions. Frankly, much work remains before this experimentation. approach bears fruit. However, the rich conceptual framework that underlies community dynamics suggests The Role of Experimentation in that many important insights will emerge and hopefully Developing “True” IPM for Grasshoppers will revitalize the basis of control and management planning. True IPM will require successful description of the above relationships in its development, and perhaps will lead to References Cited the development of “ecotechnology” based on a firm conceptual foundation. For example, here are the types of Belovsky, G. E.; Joern, A. 1995. The dominance of different regulating factors for rangeland grasshoppers. In: Cappuccino, N.; Price, questions that we must address experimentally: How do P. W., eds. Population dynamics: new approaches and syntheses. San grasshoppers compete for scarce food resources? Which Diego, CA: Academic Press: 359–386. species are the best competitors for the available food supply? What impacts do such interactions exert on the Belovsky, G. E.; Slade, J. B. 1995. Dynamics of two Montana grass- resulting grasshopper community structure? Will the land populations: relationships among weather, food abundance and food resource base change as environmental conditions intraspecific competition. Oecologia 101(3): 383–396. change and with what consequences? Are competitive Gauch, H. G., Jr. 1982. Multivariate analysis in community ecology. interactions altered in response to changing food sup- Cambridge, UK: Cambridge University Press. 298 p. plies? How important are natural enemies in deciding which grasshopper species survive and in what relative Hunter, M. D.; Ohgushi, T.; Price, P. W., eds. 1992. Effects of abundance? How do competition and predation interact resource distribution on plant-animal interactions. New York: Academic Press. 505 p. to affect grasshopper communities? How do abiotic (weather) and biotic (species-interaction) features of the Joern, A. 1979a. Feeding patterns in grasshoppers (Orthoptera: environment interact to affect grasshopper communities, Acrididae): factors influencing diet specialization. Oecologia 38: if they exert any influence at all? Results from experi- 325–347. ments to answer these and related questions will allow land managers to define explicitly the key interactions Joern, A. 1979b. Resource utilization and community structure in assemblages of arid grassland grasshoppers (Orthoptera: Acrididae). that describe the community relationships a particular Transactions American Entomological Society 150: 253–300. grasshopper infestation. Managers can then identify links that will provide the desired IPM results, or those that are Joern, A. 1982. Vegetation structure and microhabitat selection in susceptible to disruption and will lead to unwanted and grasshoppers (Orthoptera: Acrididae). Southwestern Naturalist 27: unintended results. 197–209. Joern, A. 1983. Host plant utilization by grasshoppers (Orthoptera: Final Comments Acrididae) from a sandhills prairie. Journal of Range Management 36: 793–797. Grasshopper IPM must focus on entire grasshopper assemblages, even if only a small proportion of the spe- Joern, A. 1986a. Resource utilization by a grasshopper assemblage from grassland communities. Triennial Proceedings Pan American cies are economic targets. Interactions among species Acridological Society 4: 75–100.

VII.13–5 Joern, A. 1986b. Experimental study of avian predation on coexisting grasshopper populations (Orthoptera: Acrididae) in a sandhills grassland. Oikos 46: 243–249.

Joern, A. 1992. Variable impact of avian predation on grasshopper assemblies in sandhills grassland. Oikos 64(3): 458-463.

Kemp, W. P.; Harvey, S. J.; O’Neill, K. M. 1990. Patterns of vegetation and grasshopper community composition. Oecologia 83: 299–308.

Kemp, W. P.; O’Neill, K. M. 1990. Habitat and insect biology revisited. American Entomologist 36: 44–49.

Onsager, J. A.; Henry, J. E. 1977. Comparison of five methods for estimating density of rangeland grasshoppers. Journal of Economic Entomology 70: 187–190.

Thompson, D. C. 1987. Sampling rangeland grasshoppers. In: Capinera, J. L., ed. Integrated pest management on rangeland: a shortgrass prairie perspective. Boulder, CO: Westview Press: 219–233.

VII.13–6 VII.14 Grasshopper Population Regulation

G. E. Belovsky

Factors controlling the dynamics of a population are forecasting tool for particular western regions and the often referred to as either limiting or regulating a popula- concept of weather as the driving factor in grasshopper tion (Sinclair 1989). Limiting factors operate to depress population dynamics should not be confused. a population without regard to its number; limiting factors are density independent. Regulating factors are spe- A number of general models have been developed to por- cial depressing factors that tend to bring the population to tray insect population dynamics (Southwood and Comins a specific number; to reach the specific number, the 1976, Berryman 1987). These models are generic and are depressing effect must be great when the population is not based upon specific mechanisms that operate upon much larger than the specific number and less when the the insect’s population but attempt to depict the insect’s population is below or near the specific number. Regu- population dynamics in terms of the shape of a Ricker lating factors are density dependent. curve. A Ricker curve (fig. VII.14–1) is a plot of a spe-

cies’ number (N) at time t (Nt) against its number at a Population ecologists have demonstrated that, although later time, t+1 (Nt+1). This type of population analysis is there may be a correlation between weather and popula- appropriate for insects that have a single generation each tion numbers, this correlation does not mean that weather year, which includes nearly all western rangeland grass- is the causal factor determining population dynamics or hoppers (Varley et al. 1973). Ricker curves are depic- even the most important factor—even if it is a limiting tions of population dynamics because their intersection factor (Horn 1968). In fact, it is well established that the with a reference line (Nt = Nt+1) defines the number to density-independent effects of weather on survival and which the population is being drawn by regulating factors reproduction cannot regulate populations. The effects (fig. VII.14–1). can only interact with regulating mechanisms to set population numbers because regulation requires the negative feedback of density dependent processes. Nt+1 Reference line Science’s understanding of grasshopper population dynamics has been largely built on long-standing observations that grasshopper numbers in a given year are correlated with temperature and precipitation (Joern and Gaines 1990). While these correlations provide conve- nient forecasting tools for pest managers, the correlations do not imply that weather is the causal mechanism limiting or regulating populations, nor that scientists understand grasshopper population dynamics. Furthermore, correlations between grasshopper numbers and weather, while statistically significant, are weak and are not consistent between different western rangelands with grasshopper numbers sometimes being greater in hot–dry years and sometimes greater in cool–wet years (see chap- N ter IV.8). 1 Figure VII.14–1—A simple Ricker curve relating the number of individuals starting the population in generation t (N ) to the number of Variability in the response to weather suggests that grass- t individuals produced by them to start the next generation (N ). The hopper populations may respond to other factors that are t+1 point where the reference line (Nt = Nt+1) intersects the Ricker curve is correlated with weather and not to the weather directly an equilibrium point that the population may approach. (for example, the abundance and nutritional value of food, the cover providing protection from predators, diseases, etc.). Consequently, the value of weather as a

VII.14–1 Three Relationships Important in depends upon the rate at which an individual enemy can Grasshopper Population Dynamics kill grasshoppers (functional response) and the number of enemies present (numerical response). The functional The shape of the Ricker curve depends upon the ecologi- and the numerical responses for a natural enemy fre- cal mechanisms that operate on the population and how quently increase to constant values as the density of prey they change in intensity with density. Three mechanisms increases; this phenomenon is observed in predator–prey may be particularly important for grasshoppers: (1) the systems ranging from insects and spiders to wolves and relationship between density and the probability of sur- deer. viving to the adult stage in the absence of natural enemies, (2) the relationship between density and the The implication is that as density of the grasshoppers probability that an individual is killed by a natural ene- increases, the proportion killed (probability of an indi- my, and (3) the relationship between the current year's vidual being killed) will first increase with density and density and the number of hatchlings produced for the then decrease. An example can be seen at a Palouse next year by each current female. In each case, density prairie site in western Montana for the grasshopper M. refers to the number of hatchlings per area that initiates sanguinipes where vertebrate predators, especially birds, the year’s population. I will review each of these are the principal natural enemies (fig. VII.14–2B) functions. (Belovsky and Slade 1993). Weather can modify the effects of these natural enemies. For example, cool– Density and Survival.—In the absence of natural ene- moist conditions can increase plant production, and mies, the relationship between initial grasshopper increased plant biomass enables grasshoppers to conceal hatchling density and survival determines the density of themselves from predators. But cool–wet conditions do adult females that can produce hatchlings. First, at low not always enhance grasshopper survival: they can densities, survival should be a constant proportion of the increase the virulence of some diseases. population set by weather and the nutritional value of foods because the individuals consume as much food as Density and Reproduction.—The relationship between they can potentially process. This survival is density the current year’s density of hatchlings and the hatchlings independent because it does not vary with the density of produced for the next year’s generation by each current grasshoppers present. Second, at higher densities, sur- female reflects two conditions. First, at low densities, vival becomes density dependent, as competition hatchling production per female should be constant reduces the food available per individual, and the mortal- because each female has all of the food that she can uti- ity rate increases. lize for egg production. This level of reproduction is density independent because it does not vary with the This survival relationship leads to a pattern where the density of hatchlings present. Second, at higher densi- density of adults increases as hatchling density increases ties, hatchling production per female should decline as and then becomes a constant set by the maximum adult the density of current hatchlings increases because each density that the available food can support. This relation- female acquires less and less of the available food. This ship can be seen at a Palouse prairie site in western Mon- level of reproduction is density dependent because it tana for Melanoplus sanguinipes where the addition of declines with the current density of hatchlings present. food increases survival to the adult stage (fig. VII.14–2A) This decline emerges as females acquire less and less (Belovsky and Slade 1995). Weather can increase or food because the increasing number of grasshoppers decrease food: cool–moist conditions tend to increase depletes the available food. The above pattern in repro- plant production, but tend to decrease the nutritional duction can be seen at a Palouse prairie site in western quality of the plants. Montana for M. sanguinipes where the addition of food increases reproduction (fig. VII.14–2C) (Belovsky and Density and Predation.—The relationship between the Slade 1995). Weather can increase or decrease food initial density of hatchling grasshoppers and an indi- availability. For example, cool–moist conditions tend to vidual's probability of being killed by natural enemies increase plant production but tend to decrease the nutritional quality of the plants.

VII.14–2 Adults/m2 20

15 A 10

0 01020 30 40 50 60 Hatchlings in generationt (no./m2)

Kills/100 present 5

7 3 1 B 2 6 9 1 5

0 015205 10

Eggs or hatchlings/female 50 Density independent

Hatchlings 40 Eggs

30 C 20 Density dependent

0 0 10 20 30 40 50 60

Hatchlings in generationt (no./m2)

Figure VII.14–2—The relationships between hatchling density of Melanoplus sanguinipes and (A) adult density, (B) the probability of an individual being killed by a predator, and (C) the production of eggs and hatchlings per adult female, as observed at a Palouse prairie site in western Montana. The vertical dashed lines relate the points where the probability of predation and reproduction per adult female begin to decline with hatchling density. (A and C are adapted from Belovsky and Slade [1995]. B is adapted from Belovsky and Slade [1993].)

VII.14–3 Using the Ricker Curve Nt+1 The above three relationships can be combined to con- struct a Ricker curve, which enables scientists to inte- A grate the effects of weather-induced density-independent mortality, natural enemy-caused mortality, and food resources. This integration produces three possible Ricker curve shapes, each reflecting a different dominant Enemy-limited form of population regulation.

Population Regulated Only by Natural Enemies.— This type of regulation occurs when the peak of the function relating the probability of being killed by a natural enemy occurs at a grasshopper density that is greater than the density at which hatchling production begins to decline and/or adult densities attain their maximum level. In this case, a Ricker curve emerges with a single peak or B two peaks, where the reference line intersects the Ricker curve only on the first peak (fig. VII.14–3A). This case emerges if the actions of the natural enemies (a) are so strong that grasshopper density cannot attain a level at Food-limited which competition for food occurs or (b) continue to increase as competition for food increases.

Population Regulated Only by Food Availability.— This type of regulation occurs when the peak of the function relating the probability of being killed by a natural enemy occurs at a grasshopper density that is much less than the density at which hatchling production begins to decline and/or adult densities attain their maximum level. C The Ricker curve emerges with two peaks, where the reference line intersects the Ricker curve only on the second C peak (fig. VII.14–3B). In this case, the population is capable of “escaping” the effects of natural enemies, Food- or enemy-limited because (a) the natural enemies are not very effective B and/or (b) the impact of the natural enemies rapidly diminishes as grasshopper density increases. A Population Regulated by Either Natural Enemies or Food Availability Depending Upon the Density of Hatchlings Initiating the Population.—This type of N regulation occurs when the peak of the function relating t the probability of being killed by a natural enemy occurs at a grasshopper density that is less, but not much less, Figure VII.14–3—The three Ricker curve shapes that emerge (see than the density at which hatchling production begins to text). decline and/or adult densities attain their maximum level. In this case, a Ricker curve emerges with two peaks,

VII.14–4 where the reference line intersects the Ricker curve at More than 12 years of observation of this population dis- three points (fig. VII.14–3C). closed that it has consistently been regulated by food availability, not by natural enemies (Belovsky and Slade The intersection with the first peak represents a popula- 1993, 1995). This fact suggests that the population is tion state regulated by natural enemies. The intersection near the intersection with the second peak of the Ricker with the second peak represents a population state regu- curve. Furthermore, this conclusion was expected given lated by food availability. The intersection lying between the three underlying functions measured at this site and the above two intersections defines the “watershed,” presented in figure VII.14–2. where populations initiated with densities less than this point become limited by natural enemies and with densi- What Weather Can Do ties greater than this point become limited by food availability. In this case, the population can “jump” from one A new perspective toward weather and grasshopper mode of regulation to the other depending upon the den- population regulation can be gained from the Ricker sities of hatchlings initiating a population from year to curve model by appreciating that weather can affect both year. density-independent mortality and food availability.

The picture of grasshopper population regulation Weather-induced density-independent mortality can oper- described above can be validated experimentally. From ate in conjunction with natural enemy mortality to pre- experimental (enclosed) populations established at differ- vent populations from attaining levels where food ent densities of M. sanguinipes at the Palouse prairie site availability becomes regulating. For the density- in western Montana, the Ricker curve has been measured independent mortality to be important, it would have to (fig. VII.14–4). The curve has two peaks and is inter- accomplish at least one of three things. First, inclement sected by the reference line at three points, indicating a spring weather can kill a high proportion of hatchlings, population that can be regulated by either natural enemies most likely through cold-induced starvation. Second, or food availability depending on initial hatchling weather might be sufficiently severe over the entire life densities. cycle of the grasshoppers so that few individuals can survive to become adults. Third, weather might shorten the period of time that adults have to live so that the number Hatchlings in generation t+1 of hatchlings produced is dramatically diminished. 40 On the other hand, weather exerts a far more pervasive influence by altering food availability from year to year 30 (see chapters IV.4 and IV.5). This variation in food abundance can be as great as sixfold between years and more than twofold within a summer (Belovsky and Slade 20 1995). The variation in food abundance could easily shift the shape of the Ricker curve from producing a popula- 3 3 tion regulated by natural enemies in years with low food 10 abundance to a population regulated by food abundance in years with high food abundance, and vice versa.

0 Weather Interacts With Enemies and 0102030 40 Food Availability Hatchlings in generation t

Figure VII.14–4—The Ricker curve for a M. sanguinipes population The weather-induced shifts in food abundance, and per- during a single year at a Palouse prairie site in western Montana. haps to a lesser extent, changes in density-independent Error bars and sample sizes are presented for populations initiated at mortality result in domains of attraction (shaded regions the same hatchling density.

VII.14–5 in fig. VII.14–5), where the grasshopper population fluc- tuates with weather, but is regulated by either natural Nt+1 enemies or food availability at any one time. This is the A point made by Horn (1968) that weather can create population fluctuations by varying density-independent or density-dependent (such as food availability) factors, but the density-dependent factor(s) must still regulate the population (attract it to particular levels).

In some environments, the points of attraction may be set by population levels created by natural enemies in different years (fig. VII.14–5A). In other environments, the points of attraction may be set by population levels created by food availability in different years (fig. VII.14– 5B). In still other environments, the points of attraction may vary between levels set by natural enemies in some B years and food availability in other years (fig. VII.14– 5C).

Unique spatial relationships for population regulation emerge when several populations are placed in juxtaposi- tion. The above discussion considers that each population is isolated from other populations. The conclusions concerning the regulation of a single population may have to be modified when adjacent populations are considered. For example, consider two adjacent or near populations. One population is regulated by natural enemies (fig. VII.14–3A) and the other population, by food C availability (fig. VII.14–3B). It is possible that the food- regulated population will produce individuals that migrate rather than die. Therefore, if the two populations are close enough in relation to the dispersal ability of the grasshopper, the population that would otherwise be regulated by natural enemies may be able to increase in density with the addition of immigrants and, thereby, become food regulated. The immigrants permit the population to escape the effects of natural enemies.

The above simple scenario says that in some situations Nt pest managers need to understand not only how individual populations are regulated but also the juxtaposi- Figure VII.14–5—Domains of attraction might emerge for grasshop- tion (landscape) of populations to determine the potential per population regulation, where natural enemies along with for population regulation to be complicated by dispersal. weather—which primarily affects density-independent survival and For example, the population receiving dispersers and reproduction—sets the bounds of population fluctuations (A); competition for food along with weather–which primarily affects food abun- thereby escaping regulation by natural enemies might be dance–sets the bounds of population fluctuations (B); or natural causing economic damage, and pest managers might de- enemies and food competition in different years with weather set the cide to control it. However, control of this population bounds of population fluctuations (C).

VII.14–6 might be largely ineffective unless the nearby population than 20 percent of the grasshopper nymphs—an applica- providing dispersers is controlled, too. In this scenario, tion level much less than commonly employed—could the population causing damage is not the population that shift a population from being regulated by food availabil- should be controlled because the dynamics of the former ity to being regulated by natural enemies. Switching to are dependent on the latter. such a spray regimen would lessen control costs directly and also indirectly, by taking advantage of the more The implications of population regulation for grasshopper effective actions of natural enemies. Low-mortality management may seem of little importance to managers spraying also would lead to less future management entrusted with reducing the economic damage caused by activity, with further cost reductions, because natural pest grasshoppers. However, understanding how particu- enemies would help to suppress future population lar populations entrusted to a manager are regulated can increases. provide critical insights that could make monitoring and control more cost effective. Understanding how grasshopper populations are regulated and how regulation differs between regions of the General Conclusions western rangelands is essential for the development of new control strategies that involve reduced insecticide In terms of monitoring, the following generalizations use, biocontrol agents, and grazing and habitat might be reached: manipulation.

1. Populations consistently within a domain that is regu- References Cited lated by natural enemies seldom reach densities at which economic damage is sufficient to warrant con- Belovsky, G. E.; Slade, J. B. 1993. The role of vertebrate and inverte- trol; therefore, these populations may not warrant brate predators in a grasshopper community. Oikos 68: 193–201. monitoring. Belovsky, G. E.; Slade, J. B. 1995. Dynamics of some Montana grasshopper populations: relationships among weather, food abundance and 2. Populations consistently within a domain that is regu- intraspecific competition. Oecologia 101: 383–396. lated by food availability often reach densities that cause economic damage and regularly warrant con- Berryman, A. A. 1987. The theory and classification of outbreaks. In: Barbosa, P.; Schultz, J. C., eds. Insect outbreaks. New York: trol; therefore, these populations may not warrant Academic Press: 3–30. monitoring. Horn, H. S. 1968. Regulation of animal numbers: a model counter- 3. Populations in a domain where regulation can fre- example. Ecology 49: 776–78. quently “jump” between natural enemy limitation and food limitation will only periodically cause economic Joern, A.; Gaines, S. B. 1990. Population dynamics and regulation in grasshoppers. In: Chapman, R. F.; Joern, A., eds. Biology of grasshop- damage and warrant control; therefore, these popula- pers. New York: John Wiley and Sons: 415–82. tions may warrant monitoring. Sinclair, A.R.E. 1989. Population regulation in animals. In: Cherett, If a manager knows the mode of regulation operating on J. M., ed. Ecological concepts. Oxford, UK: Blackwell Scientific: a specific grasshopper population, monitoring efforts can 197–241. be more effectively carried out, and that will save time Southwood, T.R.E.; Comins, H. N. 1976. A synoptic population and money. model. Journal of Animal Ecology 45: 949–65.

In terms of control strategies, with the knowledge of how Varley, C. G.; Gradwell, G. R.; Hassell, M. P. 1973. Insect population a population is regulated, a manager may be able to en- ecology: an analytical approach. Berkeley, CA: University of Califor- hance efficiency by creating strategies that are tailored to nia Press: 212 p. the particular population. For example, I found (1992 unpubl.) that an insecticide application that killed less

VII.14–7 Reference Cited–Unpublished

Belovsky, G. E. 1992. Grasshopper control: implications of investigations into population/community ecology. In: Grasshopper Inte- grated Pest Management Project, 1992 annual report. Boise, ID: U.S. Department of Agriculture, Animal and Plant Health Inspection Service: 57–63.

VII.14–8 VII.15 Grasshopper Habitat Manipulation

G. E. Belovsky, M. A. Brusven, D. J. Fielding, and L. Manske

Introduction Fostering Natural Enemy Abundance

Managing grasshopper populations through habitat If pest managers could change the vegetation, doing so manipulation (changes) is poorly understood and conse- might increase natural enemies of grasshopper species quently, seldom considered. However, it may be a very that reach outbreak levels. Such increases could reduce reasonable strategy given the diversity of grasshopper abundance of the pest grasshoppers and the frequency of species found in any single habitat (vegetation type) and outbreaks (Belovsky and Slade 1993). the large area that pest managers must deal with in the rangelands of the Western United States. In fact, habitat Predators as Grasshopper Population Regulators.— management, such as destruction of prime egg-laying Predators, especially vertebrates such as birds and sites, was one of the earliest and most common forms of rodents, are potentially important in regulating grasshop- grasshopper control (Pfadt and Hardy 1987). per numbers under certain circumstances (see chapter VII.14). It may be possible by habitat manipulations to Habitat manipulation would seem particularly useful extend the circumstances under which predators effec- today because many grasshopper outbreaks occur in tively limit grasshopper numbers. First, greater vegeta- habitats that have been changed by human activities. tive cover may increase the numbers of these predators Overgrazing, modified fire regimes, and introduction of by protecting rodents and bird nests from their predators. exotic plants on American rangelands have led in some Second, less vegetative cover (open vs. thick areas) can instances to replacement of relatively grasshopper- make grasshoppers more vulnerable to predators (fig. resistant native vegetation with vegetation that supports VII.15–1). The figures in this illustration were measured more frequent grasshopper outbreaks. An example may by placing tethered grasshoppers in areas of different be when the native, perennial sagebrush/bunchgrass of vegetative cover and determining how many were killed the Intermountain regions are replaced with annual by predators. grasses and forbs. Therefore, restoration of the land’s productivity can go hand in hand with grasshopper control by habitat manipulation. Percent killed The potential use of habitat manipulation as a control 3 strategy is apparent when the following two possibilities are taken into consideration: (1) Most grasshopper species do not reach outbreak levels or cause economic damage (Pfadt 1988). What if managers could replace 2 species that reach outbreak levels and cause economic damage with species that do not? Species substitution on this scale might be possible through habitat manipulation. (2) Even if outbreak species cannot be totally replaced, 1 habitat manipulations may reduce their abundance and lessen the likelihood of outbreaks.

To address these habitat manipulation prospects, we can 0 provide some potential examples but cannot present gen- Open Thick eral strategies because this issue has not been broadly examined. When we refer to habitat manipulation, we Figure VII.15–1—Comparison of the effectiveness of predators at are largely concentrating on vegetation changes because killing grasshoppers in grasslands with more than 40 percent bare both the absolute and relative abundance of grasshoppers ground (open) versus less than 20 percent bare ground (thick) in western Montana. are related to vegetation (Kemp et al. 1989, Belovsky and Slade 1995). Vegetation changes can have a variety of impacts.

VII.15–1 The effects of habitat on predation might seem in opposi- Reducing Grasshopper Food Abundance tion—on one hand increasing cover for birds and on the other hand decreasing cover for grasshoppers. However, In many areas of western rangeland, food abundance may on rangelands, the management trend is to make them be limiting grasshopper populations (see chapter VII.14). more uniform. For example, overgrazing tends to reduce It may be possible to diminish food abundance using the height of vegetation; while this factor can make the habitat manipulations in ways that will not negatively grasshoppers more vulnerable to predation, there are now affect the forage available to livestock. fewer predators to take advantage of the more open conditions for hunting, so the potential for greater predation Increasing Competitors’ Abundance.—If other species on grasshoppers is seldom fully realized. compete with the pest grasshoppers for food, then increasing the abundance of these competitors might Manipulation might restore some of the natural variation reduce the abundance of pest grasshoppers. Unfortu- in the habitat. Changes of that sort might be accom- nately, enhancing the numbers of competitors might sim- plished by providing small patches of thick cover for pro- ply substitute one pest for another so that the forage tection of the grasshoppers’ predators, especially available to livestock is not enhanced. However, limiting bird-nesting sites. Simultaneously, a pest manager might pest grasshoppers by reducing their available food maintain habitat openness or even reduce cover in the through consumption by competitors, without simulta- intervening larger areas between patches of thick cover to neously diminishing the forage available to livestock, increase the effectiveness of the predators in capturing might be accomplished under two conditions. First, live- grasshoppers. In doing this, a manager might be able to stock grazing might be used to reduce grasshopper num- increase the predators’ numbers and efficiency and bers; this substitutes livestock consumption for grasshop- thereby enhance the ability of predators to limit grasshopper consumption of the forage. Second, habitat manipu- pers when predators otherwise might not be effective. lations might be used to replace pest grasshopper species with species that do not reach outbreak levels, especially Parasitoids and Parasites.—As with predators, parasi- if these other species do not reduce the forage for live- toids and parasites might have their numbers and effi- stock to as great a degree as the pest species. ciency enhanced by manipulating the vegetation. For example, mites (parasites that attach themselves to a Different studies have disclosed that livestock grazing grasshopper’s exoskeleton and “suck” the grasshopper’s decreases grasshopper densities (Hutchinson and King “blood”) can dramatically reduce grasshopper survival 1980; Jepson–Innes and Bock 1989; Capinera and and egg production, but these parasites generally do not Sechrist 1982; Fielding and Brusven 1995), increases appear to reach high enough densities to limit grasshop- densities (Coyner 1938 unpubl., Nerney 1958, Anderson pers (see chapter I.9). 1964, Holmes et al. 1979), and has no effect (Miller and Onsager 1991) on grasshopper densities. In cases where The inability of mites to reach high enough densities to grazing reduced grasshopper abundance, it appeared that limit grasshopper populations appears to be due in many the grasshoppers encountered a shortage of food. In areas to soils that have reduced drainage. Poor drainage cases where grazing increased grasshopper abundance, it should not be confused with moist conditions, a rarity in appeared that the grasshoppers either responded to most western rangelands; poor drainage pertains to soils, decreased cover (see thermal cover, below) or increased such as clays, that tend to hold moisture longer. As with forb abundance (see vegetation changes, below). All of cover for predators, a manager might consider creating the above studies found that the grasshopper species patches favorable to mite production that are interspersed composition changed with grazing. Grazing effects are throughout the larger area. Changing vegetation compo- more fully discussed in chapter V.1. sition or cover or even providing small areas of better draining soils in small areas could achieve this end.

VII.15–2 Grasshoppers that compete with the pest species might Number/cage be encouraged by management to reduce the pests’ abun- 1.5 dance. This option would be useful if the competitor emerges earlier than the pest, so that survival of the pest species’ nymphs is reduced. In addition, it would be particularly useful if the earlier emerging competitor cannot survive later into the season, when the pest would 1 otherwise be most abundant; this scenario would allow the vegetation to regrow after consumption by the competitor. 0.5 An example is provided by the nonpest early-season grasshopper Melanoplus confusus and the pest late- season grasshopper, M. sanguinipes, in the Palouse prairie of western Montana (Belovsky 1990 unpubl). As 0 fourth- and fifth-instar nymphs and adults, M. confusus Alone Together Added dramatically reduces the survival of M. sanguinipes in experimental populations by competing for food plants Figure VII.15–2—The densities attained by Melanoplus sanguinipes (fig. VII.15–2). The M. confusus adults quickly die off in in experimental field populations (cages) when by itself (Alone), when with M. confusus (Together), and when it is added after M. confusus early July, and the vegetation regrows because rains in dies off later in the summer (Added). most years permit continued growth. The negative effect of M. confusus on M. sanguinipes is illustrated by M. sanguinipes being able to reach the same densities in the experimental mixed populations as in experimental Therefore, habitat manipulations that modify the relative pure populations, when M. sanguinipes are placed in the abundances of plants need to be weighed against changes experiments after M. confusus dies off (fig. VII.15–2). in these other factors and how they affect both pest and Unfortunately, under natural conditions, M. confusus livestock. populations are generally too low to achieve this effect. Changing Grasshopper Thermal Cover Encouraging M. confusus.—A straightforward means by which a manager might increase M. confusus numbers Vegetation provides more than food—it also provides is not apparent. thermal cover for grasshoppers. Grasshoppers are able to consume a greater quantity of food when they are in Manipulating Plant Species.—The relative abundance favorable thermal conditions. Under favorable condi- of different plant species might be manipulated to reduce tions, a grasshopper can process more food through its the abundance of those species that are more important to digestive tract and has more time to consume foods. the pest grasshoppers than they are to livestock. While Greater food consumption leads not only to greater grasshoppers and livestock consume many of the same immediate losses of forage resources on rangelands but plant species and thereby compete, grasshoppers do not also to larger grasshopper populations by increasing the consume identical sets of food plants. A good example grasshoppers’ survival and reproduction. of this manipulation might be to reduce the abundance of annual grasses and forbs and to increase the abundance of Thick vegetative cover for a grasshopper may lead to perennial grasses and shrubs. Many pest grasshoppers, a thermal environment that is cooler than optimal, especially in the spurthroated group (Melanoplines), reducing grasshopper survival and reproduction. The seem to thrive with the annuals, and livestock are capable same effect can be caused when there is too little vegeta- of foraging on the perennials. But changing vegetative tive cover for a grasshopper and the environment is composition can also modify cover and plant abundance. warmer than optimal. Therefore, land managers might

VII.15–3 manipulate vegetative cover to diminish grasshopper A greater variety of these methods needs to be investi- feeding, and thereby, their survival and reproduction. gated in a range of different habitats. However, these methods may require greater than normal monitoring by Possible Methods for Habitat managers. For example, grazing and fire both require the Manipulation manager to assess intensity carefully, and doing that can be difficult as weather conditions dramatically change the We have presented a series of ecological processes that vegetation from year to year. habitat management might be able to exploit to reduce pest grasshoppers. However, methods are required to For example, management by grazing might require the modify the habitat and thereby change the ecological manager to manipulate stocking rates much more than processes. ranchers traditionally have undertaken, or in ways that do not maximize the rancher’s income. In addition, habitat A number of methods have been investigated without ref- manipulations must be evaluated in terms of their impacts erence to how they changed ecological processes. It has on wildlife, recreation activities, and the maintenance and been demonstrated that the use of herbicides on range- restoration of native vegetation. Habitat manipulations lands has little effect on grasshoppers, while furrowing, have not been adequately investigated as a viable pest- scalping, and interseeding grazing lands can reduce management strategy for grasshoppers, but manipulations grasshopper numbers dramatically (Hewitt and Rees may have great potential to reduce grasshopper-caused 1974). Researchers are not sure if furrowing, scalping, damage with fewer negative impacts on the environment. and interseeding change predation cover, thermal cover, plant composition, or all of these factors. References Cited

One method that has been investigated at least partially Anderson, N. L. 1964. Some relationships between grasshoppers and vegetation. Annals of Entomological Society of America 57: 736–42. from the perspective of ecological processes operating on pest grasshoppers is fire on rangelands. It primarily oper- Belovsky, G. E.; Slade, J. B. 1993. The role of vertebrate and inverte- ates to change the composition of the vegetation and, brate predators in a grasshopper community. Oikos 68: 193–201. thereby, grasshopper food abundance. However, fire can produce different outcomes on pest grasshoppers. Under Belovsky, G. E.; Slade, J. B. 1995. Dynamics of some Montana grass- some conditions, fire enhances grasshopper numbers and hopper populations: relationships among weather, food abundance and intraspecific competition. Oecologia 101: 383–396. in others, decreases them. For example, intense fires destroy sagebrush/native bunchgrasses, enhancing annual Capinera, J. L.; Sechrist, T. S. 1982. Grasshopper (Acrididae)–host plants, which are favored by pest grasshoppers. On the plant associations: response of grasshopper populations to cattle graz- other hand, “cool” fires enhance the abundance of native ing intensity. Canadian Entomologist 114: 1055–62. bunchgrasses and, thereby, decrease pest grasshoppers. Fielding, D. J.; Brusven, M. A. 1995. Grasshopper densities on grazed Likewise, livestock grazing can be used to manipulate and ungrazed rangeland under drought conditions in southern Idaho. vegetation composition, but as with fire, different grazing Great Basin Naturalist 55:352–358. intensities result in different outcomes. Hewitt, G. B.; Rees, N. E. 1974. Abundance of grasshoppers in rela- Reseeding areas with crested wheatgrass after native tion to rangeland renovation practices. Journal of Range Management bunchgrasses have been destroyed can reduce pest grass- 27: 156–60. hopper abundance but not to the extent that native bunch- Holmes, N. D.; Smith, D. S.; Johnston, A. 1979. Effect grasses can. Therefore, methods for restoring native of grazing by cattle on the abundance of grasshoppers rangelands may have considerable potential for on fescue grassland. Journal of Range Management 32: 310–11. grasshopper pest management.

VII.15–4 Hutchinson, K. L.; King, K. L. 1980. The effects of sheep stocking Pfadt, R. E. 1988–1997. Field guide to common western grasshoppers. level on invertebrate abundance, biomass and energy utilization in a Bull. 912. Laramie, WY: University of Wyoming and Wyoming Agri- temperate, sown grassland. Journal of Applied Ecology 17: 369–87. cultural Experiment Station. 250 p.

Jepson–Innes, K.; Bock, C. E. 1989. Response of grasshoppers Pfadt, R. E.; Hardy, D. M. 1987. A historical look at rangeland grass- (Orthoptera: Acrididae) to livestock grazing in southeastern Arizona: hoppers and the value of grasshopper control programs. In: Capinera, differences between seasons and subfamilies. Oecologia 78: 430–31. J. L., ed. Integrated pest management on rangeland: a shortgrass prairie perspective. Boulder CO: Westview Press: 183–95. Kemp, W. P.; Kalaris, T. M.; Quimby, W. F. 1989. Rangeland grasshopper (Orthoptera: Acrididae) spatial variability: macroscale popula- References Cited–Unpublished tion assessment. Journal of Economic Entomology 82: 1270–76. Belovsky, G. E. 1990. Grasshopper competition and predation: bio- Miller, R. H.; Onsager, J. A. 1991. Grasshopper (Orthoptera: logical control options. In: Grasshopper Integrated Pest Management Acrididae) and plant relationships under different grazing intensities. Project, 1990 annual report. Boise, ID: U.S. Department of Agricul- Environmental Entomology 20: ture, Animal and Plant Health Inspection Service: 37–44. 807–14. Coyner, W. R. 1938. Insect distribution and seasonal succession in Nerney, N. J. 1958. Grasshopper infestations in relation to range overgrazed and normal grasslands. Unpubl. M.S. thesis. Norman, OK: condition. Journal of Range Management University of Oklahoma. 78 p. 11: 247.

VII.15–5

VII.16 Grasshoppers—Plus and Minus: The Grasshopper Problem on a Regional Basis and a Look at Beneficial Effects of Grasshoppers

G. E. Belovsky, A. Joern, and J. Lockwood

Introduction needed for livestock and wildlife. Of course, all forage that grasshoppers eat cannot be consumed by livestock From an environmental perspective, grasshopper control and wildlife. Grasshoppers have an important role in the in rangelands of the Western United States poses several ecological processes that make U.S. rangelands so pro- unique and difficult problems compared to the control of ductive. Shifting the management viewpoint from elimi- many other insect pests. nation to suppression is a difficult undertaking but places grasshopper management within the larger context of sus- • When scientists or land managers speak of grasshop- tainable ecosystem management and the preservation of pers, they are not referring to a single pest species but biodiversity. to a group of insects that contains more than 400 species, with as many as 30 to 40 species found in any Given past concern over grasshopper damage to range- given area. Some of these species cause economic land production, one would think that the scientific abil- damage, but most do not; however, current control ity to address the central issues would be much more methods influence all (Lockwood 1993a and b, extensive than it is. Most efforts have focused on con- Carruthers and Onsager 1993). trol, and perhaps in some cases eradication, of grasshoppers. With the development of commercially produced • None of these insects has been introduced to the West synthetic pesticides in the 1930’s, this focus led to a pre- by humans. All are natural elements of a complex dominance of studies intended to produce better insecti- ecological system that is highly productive for live- cides and means of application. Such a focus also stock and wildlife. Therefore, grasshoppers are an replaced investigating grasshopper biology in ways that important consideration in conservation planning might form a basis for alternate approaches. (Lockwood 1993a and b, Carruthers and Onsager 1993). An integrated pest management approach must be founded upon the biology of the pest species. The Grass- • While managers often consider rangelands to be uni- hopper Integrated Pest Management Project has helped form grasslands, rangelands can refer to mountain provide us with more information on grasshopper control meadows, savannas, forested parklands and and biology. Project-funded investigators have identified shrublands, and steppe grasslands. Rangelands vary many important questions that a pest manager must con- dramatically in plant species composition; the sider. Considering such questions is the critical first step amount, frequency, and annual distribution of precipi- in fostering the development of management strategies tation; and forage production. for particular rangeland locations in the future.

Seeking or expecting a single control strategy for pest Grasshopper Management Over the grasshoppers may be fruitless. Grasshoppers form a di- Variety of Rangelands verse group of species that inhabit a diverse group of habitats. Advocating the elimination or dramatic reduc- One simple observation from grasshopper studies illus- tion in grasshopper numbers, even if this action were bio- trates the enormous task posed by grasshopper manage- logically and economically feasible, could be destructive ment over the range of species and habitats found in the to the very ecological system whose production we are Western United States. In the southern rangelands, trying to maintain and exploit (Lockwood 1993a and b, increased precipitation and possibly cooler temperatures Mitchell and Pfadt 1974). Consequently, control may not appear to increase grasshopper numbers. In northern be a desirable goal. Management may be the more rangelands, the opposite conditions (warm and dry) appropriate perspective. appear to increase grasshopper numbers (Capinera and Horton 1989). This comparison covers an immense Grasshopper management should attempt to minimize region and glosses over the variability in vegetation competition for forage between grasshoppers, livestock, among different areas. There also are other ecological and wildlife in cases when most rangeland production is factors that lead to variation in grasshopper numbers and

VII.16–1 species composition (Joern and Gaines 1990). Further- more, we have little idea of what particular mechanisms The Ecological Role of Grasshoppers are driving the above patterns (including changes in plant production, plant nutritional value, grasshopper develop- Grasshoppers play an important role in the functioning of mental rate, predation rate, fungal infection rate, and rangeland ecosystems (Mitchell and Pfadt 1974). First, more), because the weather variables are no more than results from a variety of studies reveal that grasshoppers correlates with grasshopper numbers (Joern and Gaines typically consume at least 10 percent of available plant 1990). biomass. Second, grasshoppers often harvest more plant biomass than they consume, influencing the availability To illustrate further the problems arising from the diver- and distribution of litter in the environment. This con- sity of rangeland habitats, there are two other major dif- sumption and harvesting could be deemed negative from ferences that emerge in comparisons of southern and the perspective of available plant biomass for livestock northern rangelands. In the South, warm-season grasses production. But such “harvesting” processes can serve dominate, and the smaller bodied, slantfaced important functions for the cycling of nutrients. (Gomphocerinae) grasshoppers are most abundant. In northern areas, cool-season grasses dominate, and the Microbes can break down the feces produced by grass- larger bodied, spurthroated (Melanoplinae) and hoppers more easily than those produced by larger herbi- bandwinged (Oedopodinae) grasshoppers are most abun- vores, such as cattle or sheep. Grasshopper-generated dant. Warm-season grasses generally are less nutritious fecal nutrients are therefore more available for plant pro- for grasshoppers than cool-season grasses. Slantfaced duction. Also grasshoppers have a shorter lifespan and grasshoppers that dominate in areas with warm-season generally decompose where they die. The nutrients in grasses are better at feeding on these plants. Therefore, their bodies return more rapidly to the soil for plant use the weather correlates observed over the rangelands of than do nutrients found in the bodies of livestock. Even the Western United States are further complicated by when grasshoppers create litter, they are enhancing plant major changes in vegetation and grasshopper species production because increased litter increases the water composition. retention of soils and reduces summer soil temperatures. These phenomena, in turn, enhance plant production by The above points illustrate the need to better define the making more water and nutrients available in the semi- environmental conditions that affect grasshoppers in dif- arid and arid conditions of the West. In total, grasshop- ferent regions and the ways that grasshopper populations pers may exert a positive influence on rangeland plant function. Furthermore, some evidence suggests that production. rapid, human-induced climate changes could make iden- tifying regional patterns worth little to managers. Cli- Grasshoppers selectively feed on different plant species mate changes may produce new patterns rather than and, consequently, influence the plant species composi- simple latitudinal displacements of existing patterns tion of the ecosystem. Sometimes, the grasshoppers har- (southern rangelands may not simply move northward). vest plants that livestock prefer. In other instances, Similarly, other human-induced changes in the environ- grasshoppers consume plants that are poisonous or com- ment (changes in the abundances of native plant species petitively reduce the abundance of plants preferred by and introductions of exotic plants and animals) could dis- livestock. The selective consumption of different plant rupt observed patterns. Therefore, people need to under- species by grasshoppers can change the nutrient cycling stand the different processes creating the patterns dynamics in a rangeland. This change happens because observed in different western U.S. rangelands. By doing the total nutrient content and decomposition rate of the so, managers can anticipate and plan responses to the litter depend on the plant species composing the litter changing environments, policies and values that will con- (Pastor et al. 1987). Therefore, selective consumption of front us in the future. certain plant species can have a positive or negative effect on primary production for livestock by changing plant species abundances and nutrient cycling.

VII.16–2 Grasshoppers are a major food source for other species reduce the abundance of prickly pear cacti. Even more that inhabit rangelands, especially spiders, reptiles, birds, important, grasshoppers may prevent or retard the spread and small mammals. Consequently, grasshoppers sup- of exotic weeds, as with feeding by Aeoloplides turnbulli port other biological components of the ecosystem and and Melanoplus lakinus on Russian thistle (Salsola influence their ability to affect ecosystem functioning. iberica). Scientists need to investigate more fully the Again, grasshoppers can positively or negatively influ- potential benefit of weed control through grasshopper ence the biological composition of ecosystems and their feeding. This area of research could become especially productivity for livestock. important with the difficult problem of controlling the spread of exotic weeds on rangelands. Weeds compete With the increasing emphasis placed upon ecosystem with native flora, and livestock find many weeds management by Federal and State agencies, grasshoppers especially unpalatable. in the rangelands of the Western United States must be considered in terms of their beneficial actions, not just in Grasshoppers and Conservation terms of their potential to reduce the abundance of forage for livestock. Consequently, pest management cannot be Clearly grasshoppers can provide many benefits that the considered in isolation from larger ecological issues. public frequently has overlooked for the conservation of This is especially true when the pest is a natural, rangelands. In addition, there is growing social and coevolved component of the ecosystem, as grasshoppers political concern for the protection of biodiversity. Con- are in western rangelands. Land managers must explic- cern increases because of unrecognized benefits provided itly acknowledge that in most years, in most places, most by many species and their important role in maintaining grasshopper species do not harm the rangeland resource; healthy ecosystems, and because these species are an rather they may benefit the resource. important part of our cultural history and they are estheti- cally pleasing (Wilson 1989). Finally, there is a growing Grasshoppers as a Range-Management view in U.S. society that people have an ethical obliga- Tool tion to ensure the continued existence of all species and the ecosystems that they inhabit. The view is that each Considering the important role grasshoppers serve in eco- species has the same evolutionary value as the human systems, these insects deserve consideration as a tool species, and ecosystems have the same value as human land managers could employ to enhance rangeland pro- society (Kellert and Wilson 1993). ductivity for livestock. First, nutrient cycling must be maintained to preserve or enhance rangeland production, Grasshoppers usually are abundant enough to be exempt and grasshoppers may aid in this goal. Second, the selec- from threats of extinction. Nonetheless, at least one spe- tive foraging of grasshoppers on different plant species cies of grasshopper that was a very abundant pest appears might increase the abundance of plants that are more pal- to have become extinct, the Rocky Mountain locust atable and beneficial to livestock. Therefore, the nega- (Melanoplus spretus). This species did not die out from tive effects of grasshoppers on forage availability for control efforts but probably from habitat destruction livestock must be compared against their positive effects caused by agriculture and livestock grazing (Lockwood on maintaining or enhancing rangelands. and DeBrey 1990).

Perhaps the greatest potential of grasshoppers as a man- Not many years ago, the loss of the Rocky Mountain agement tool may be to alleviate the growing problem of locust was considered a benefit. Today, many view this weed control (Lockwood 1993a). For example, it loss with apprehension. Few people would wish a return appears that the grasshopper Hesperotettix viridis may to the state where this species destroyed croplands, but control the abundance and spread of snakeweed the public can no longer experience, even on a small (Gutierrezia spp.), rabbitbrush (Chrysothamnus spp.), scale, the swarms that darkened the skies and stopped ragweed (Ambrosia spp.), and locoweeds (Astralagus transcontinental railroads as told as part of America’s spp.). The grasshopper Melanoplus occidentalis may national heritage and folklore. More importantly, the loss

VII.16–3 of the Rocky Mountain locust means that an important Questions for the Future element of the Nation’s pristine rangelands has been lost, and the loss exemplifies the general assault upon natural One certainty for the future is that grasshopper manage- environments, especially rangelands, by human actions. ment will be changing. There will be little “business as usual.” For example, exotic plant species have almost entirely replaced the native annual grasslands of California. Only • The methods of grasshopper control will change as remnants of tallgrass prairie remain, and the introduction society becomes more concerned with environmental of exotic plants threatens most other western rangelands. degradation and the protection of all native species. What will happen to the native grasshoppers that inhabit Therefore, new and innovative control methods that these ecosystems? Several species of monkey grasshop- are environmentally sound will need to be found and pers in native desert grasslands are considered threatened used. and may eventually be listed for protection under the En- dangered Species Act. • Grasshoppers, as native components of rangelands, will no longer be considered solely as pests to be sup- The decline of grasshoppers also affects other species, pressed or eradicated, but as important elements for especially those that consume them. Recently, the U.S. the functioning of our natural ecosystems. Further- Fish and Wildlife Service announced that western range- more, society is beginning to view all species that are land birds have dramatically declined in abundance over part of our native biodiversity as having esthetic the last decade, with the numbers of some species value, as providing a reflection of our national heri- decreasing by as much as 70 percent. Many of these tage that deserves some level of protection, and as birds feed on grasshoppers as adults, and almost all rely requiring protection from an ethical perspective. The heavily on grasshoppers to provision their nestlings. short-term economic costs/benefits of pest control to Therefore, the control of grasshoppers must be consid- livestock production will become less important in ered in a broader conservation perspective than forage decisionmaking and more subject to review by production for livestock, protection of threatened grass- society. hopper species, and the maintenance of the ecosystem functions provided by grasshoppers. Grasshopper reduc- • The general patterns of grasshopper abundance in dif- tion also might harm declining or threatened species that ferent regions will change if humans change the glo- depend on these insects as food (Belovsky 1993). bal climate as projected by many scientists. Therefore, managers must act in places and ways pre- Conservation concerns are becoming more pronounced in viously unanticipated. The result is that pest manag- formulating management plans because of legal and ers need to adopt a broader perspective of their role, social mandates. Therefore, the scope and scale of grass- become more flexible in their actions, and view the hopper control programs will no doubt become more changing environment as an exciting challenge, rather restricted in the future and will require consideration of than a hindrance. far more than the short-term economic costs of grasshopper consumption of livestock forage.

VII.16–4 References Cited Lockwood, J. A. 1993a. Environmental issues involved in biological control of rangeland grasshoppers (Orthoptera: Acrididae) with exotic Belovsky, G. E. 1993. Modelling avian foraging: implications for agents. Environmental Entomology 22: 503–518. assessing the ecological effects of pesticides. In: Kendall, R. J.; Lacher, T. E., Jr., eds. Wildlife toxicology and population modeling: Lockwood, J. A. 1993b. Benefits and costs of controlling rangeland integrated studies of agroecosystems. Boca Raton, FL: CRC Press: grasshoppers (Orthoptera: Acrididae) with exotic organisms: search 123–138. for a null hypothesis and regulatory compromise. Environmental Entomology 22: 904–914. Capinera, J. L.; Horton, D. R. 1989. Geographic variation in effects of weather on grasshopper populations. Environmental Entomology 18: Lockwood, J.; DeBrey, L. D. 1990. A solution for the sudden and 8–14. unexplained extinction of the Rocky Mountain locust, Melanoplus spretus (Walsh). Environmental Entomology 19: 1194–1205. Carruthers, R. I.; Onsager, J. A. 1993. Perspective on the use of exotic natural enemies for biological control of pest grasshoppers (Orthop- Mitchell, J. E.; Pfadt, R. E. 1974. The role of grasshoppers in a short- tera: Acrididae). Environmental Entomology 22: 885–903. grass prairie ecosystem. Environmental Entomology 3: 358–360.

Joern, A.; Gaines, S. B. 1990. Population dynamics and regulation in Pastor, J.; Dewey, B.; Naiman, R. J.; McInnes, P. 1987. Moose, grasshoppers. In: Chapman, R. F.; Joern, A., eds. Biology of grasshop- microbes and the boreal forest. BioScience 38: 770–777. pers. New York: John Wiley and Sons: 415–483. Wilson, E. O. 1990. Threats to biodiversity. Scientific American 9: Kellert, S. R.; Wilson, E. O. 1993. The biophilia hypothesis. 108–116. Washington, DC: Island Press. 484 p.

VII.16–5